Ultrasound features as predictive markers of BRAFV600E mutation in thyroid cancer: a systematic review and meta-analysis

Introduction

The past few years have witnessed a considerable increase in thyroid carcinoma cases (1,2). Papillary thyroid carcinoma (PTC), the most prevalent histological variant, constitutes about 80–90% of all thyroid cancer instances, and its incidence continues to increase annually (3). While the majority of thyroid cancers are associated with a favorable prognosis and low mortality rates, a minority of high-risk patients face extrathyroidal invasion, recurrence, and metastasis, especially in advanced stages. Thus, accurate preoperative evaluation of the lesion’s characteristics and its invasiveness is critical for surgical decision-making and planning.

Earlier research indicates that the BRAFV600E mutation can serve as a valuable biomarker and therapeutic target for the diagnosis, risk stratification, and prognosis prediction of PTC (4,5). Of note, the BRAFV600E mutation plays a crucial role in regulating cell proliferation, differentiation, and apoptosis (6). Moreover, it is frequently observed in thyroid carcinoma cases and is connected with PTC.

Hence, preoperative prediction of the BRAFV600E mutation holds significant clinical value for the diagnosis and treatment of PTC. Previous studies have actively explored preoperative prediction of the presence of BRAFV600E mutations in thyroid cancer based on conventional ultrasound imaging (7). Some studies reported that conventional ultrasound features could predict BRAFV600E mutations in PTC (8-14), whereas others have reached opposite conclusions (7,15,16), leading to significant discrepancies in the field. Currently, there are no published meta-analyses on the use of ultrasound features to predict BRAFV600E mutations preoperatively in thyroid cancer patients, which are significant prognostic factors and would assist in the diagnosis and differentiation of the disease.

Therefore, this study aimed to conduct a meta-analysis to determine the predictive value of ultrasound features for BRAFV600E gene expression status preoperatively, laying a theoretical reference to potentially mitigate unnecessary invasive procedures such as fine-needle biopsies. We present this article in accordance with the PRISMA reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-24-134/rc).

Methods

Literature search

Literature searches were conducted in PubMed, Web of Science, and Cochrane Library using the terms “thyroid cancer” and “ultrasonography”, or “sonography”, or “ultrasonic”, or “ultrasound”, and “BRAF” until December 2023.

A total of 214 articles was yielded by our search algorithm: 138 from PubMed, 74 from Web of Science, and 2 from Cochrane Library. EndNote X8 was used for literature management, and duplicates were manually excluded.

The inclusion criteria for this study were: (I) research on the ultrasound characteristics of thyroid cancer; (II) histopathology or fine needle aspiration (FNA) cytology as the standard reference; (III) evaluation of ultrasound features for predicting BRAFV600E mutation status; and (IV) studies with retrospective or prospective designs.

Exclusion criteria were: (I) studies not focusing on thyroid cancer; (II) unavailable data in 2×2 diagnostic tables; and (III) studies containing duplicated data or instances where patients may have overlapped across studies.

Data extraction

Data extracted from the selected studies comprised (I) study details such as first author, publication year, design, and inclusion period; and (II) patient demographics such as gender, age, and number of malignant nodules. Two reviewers independently screened the literature and extracted data into a standardized Microsoft Excel spreadsheet to ensure consistency.

Literature evaluation criteria

Study quality was assessed using the Quality Assessment Tool for Diagnostic Accuracy Studies-2 (QUADAS-2) tool. Each article was evaluated for quality and categorized as “yes”, “no”, or “unclear”. Literature was rated “yes” if the criteria were met, “no” if the criteria were unmet or unspecified, and “unclear” if the provided information was incomplete.

Furthermore, the included studies underwent risk of bias and applicability evaluation relative to the research question, with “low”, “high”, or “unclear” ratings denoting their applicability and bias risk levels. Disagreements were resolved via discussion.

Statistical analysis

Statistical analyses were performed using RevMan 5.4 and Stata 12.0. The risk of bias in the included studies was assessed using bias risk graphs. The sensitivity and specificity of different ultrasound features in predicting the expression status of the BRAFV600E gene in thyroid cancer were analyzed, and the summary receiver operating characteristic (SROC) curves were plotted. The area under the curve (AUC) was calculated with 95% confidence intervals (CIs). Heterogeneousness among studies was evaluated using the Q test and I² statistics. An I² inconsistency index was calculated to determine heterogeneity’s level; high heterogeneousness was indicated by an I²≥50%, and the random-effects model was adopted. In heterogeneity’s absence, the fixed-effects model was applied. Funnel plots with different diagnostic indicators were used to detect potential publication bias, and sensitivity analyses were conducted. Forest plots generated by the software provided 95% CIs and P values for evaluating meta-analysis outcomes, with P<0.05 considered statistically significant.

Results

Literature search results and characteristics of included studies

The extensive search across the three databases yielded 214 relevant articles. Among them, 68 were found to be duplicates and thus excluded. A detailed review of the titles and abstracts led to the exclusion of 116 articles for their lack of relevance. Further scrutiny of the full texts resulted in the exclusion of 15 additional articles due to their insufficient relevance to the research topic. Moreover, two articles were excluded during the full-text assessment, given that it was not feasible to derive a 2×2 table from the data. Ultimately, 13 articles met the criteria and were incorporated into the systematic review and meta-analysis (8-20) (Figure 1).

Figure 1 PRISMA flowchart of the study selection process. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-analyses.

This study consisted of 13 studies involving 2,250 patients with thyroid cancer, published between 2012 and 2022. Of these, eight were retrospective, three were prospective, and two were unspecified. The research was conducted across four countries: Poland, the United States, South Korea, and China. Sample sizes varied from 34 to 438 patients, with BRAFV600E mutation rates ranging from 47.7% to 86.5%. Notably, lower mutation rates were observed in European and American countries, while higher rates were found in East Asian countries (Table 1).

Quality assessment

Two reviewers independently conducted the quality assessment of the articles included in the study. In case of discrepancies between their evaluations, the decision was made following a thorough discussion. Based on the QUADAS-2 questionnaire outcomes, each study was characterized by a low risk of bias and qualified as high quality. This assessment is clearly reflected in both the literature risk of bias evaluation graph and the summary graph in Figure 2.

Figure 2 Outcomes of QUADAS-2 for included studies. (A) Risk-of-bias summary. (B) Risk-of-bias graph. Symbols: (+), low risk of bias; (?), unclear risk of bias; (−), high risk of bias. QUADAS-2, Quality Assessment Tool for Diagnostic Accuracy Studies-2.

Evaluation of the diagnostic performance of ultrasound features

Notably, the combined sensitivity of the seven ultrasonographic features (as shown in Appendix 1) for predicting BRAFV600E gene expression in thyroid cancer was satisfactory. The ultrasonographic features with high combined sensitivity included hypoechogenicity, solid composition, absence of halo, and ill-defined borders, with sensitivities of 0.93 (95% CI: 0.87–0.96), 0.93 (95% CI: 0.86–0.97), 0.83 (95% CI: 0.72–0.91), and 0.74 (95% CI: 0.64–0.83), respectively (Figure 3 and Table 2). Additionally, the SROC curve displayed that features with high AUC were the absence of halo and hypoechogenicity, with AUC values of 0.84 (95% CI: 0.80–0.87) and 0.81 (95% CI: 0.77–0.84), respectively (Figure 4).

Figure 3 Forest plot of sensitivity and specificity among studies. (A) Hypoechoic; (B) solid composition; (C) absent halo; (D) ill-defined margin. Each horizontal line represents an individual study with the result plotted as a box and the 95% CI displayed as the line. The diamond at the bottom of each plot shows the result of the individual studies combined and averaged. The horizontal lines of the diamond are the limits of the 95% CIs. CI, confidence interval.

Figure 4 SROC curve with AUC of two sonographic features in predicting BRAF gene expression status in thyroid cancer. (A) Absent halo; (B) hypoechoic. SROC, summary receiver operator characteristic; SENS, sensitivity; SPEC, specificity; AUC, area under the curve.

Subgroup analysis and sensitivity analysis

High heterogeneousness was displayed by the forest plot in the pooled specificity and sensitivity of ultrasonographic features for predicting BRAFV600E gene expression in thyroid cancer. To address heterogeneousness and assess the results’ robustness, the random-effects model was used to calculate combined outcomes, whilst subgroup analysis was conducted to explore heterogeneity’s sources. The heterogeneity was addressed and homogenized following the exclusion of the study undertaken by Wang et al. using a leave-one-out approach (12), suggesting this study might have been a potential source of heterogeneity. Subgroup analysis of the seven ultrasonic features revealed that hypoechogenicity, absence of halo, ill-defined margins, and vertical orientation were statistically significant (P<0.05) predictors of BRAFV600E gene expression in thyroid cancer, with odds ratio (OR) values ranging from 1.02 to 36.91. However, solid nodule composition, irregular shape, and microcalcifications were not statistically significant (P>0.05) (Figure 5).

Figure 5 Odds ratio and its 95% confidence intervals of ultrasound features in predicting BRAF gene expression status in thyroid cancer. M-H, Mantel-Haenszel; CI, confidence interval.

Publication bias

Additionally, the Deeks funnel plot analysis revealed no significant publication bias across the ultrasound characteristics (P>0.05) (Figure 6).

Figure 6 Deeks funnel plot to assess the publication of bias. ESS, effective sample size; sqrt, square root.

Discussion

To the best of our knowledge, this is the first meta-analysis summarizing the prognostic value of ultrasound features in preoperatively predicting BRAFV600E gene expression status in thyroid cancer. Prior research (21-23) documented that BRAFV600E gene expression is a critical prognostic factor in thyroid cancer, especially in advanced stages. Hence, this systematic review and meta-analysis were carried out to evaluate the predictive value of ultrasound characteristics. Based on the results of the included studies, the absence of halo and hypoechogenicity in thyroid nodules exhibited superior predictive efficacy for preoperative BRAFV600E gene expression status in thyroid carcinoma compared to the remaining ultrasonographic features.

In this meta-analysis, we included 13 studies involving 2,250 patients with thyroid carcinoma to examine the value of ultrasonographic features in predicting the BRAFV600E gene expression status in thyroid carcinoma. Herein, a SROC curve was constructed for each ultrasound feature, and the AUC value was used as an indicator of the diagnostic efficacy. A larger AUC corresponds to a greater diagnostic accuracy. The findings revealed that the AUC for the absence of a halo and hypoechogenicity was notably high, at 0.84 (95% CI: 0.80–0.87) and 0.81 (95% CI: 0.77–0.84), respectively. The findings implied that the absence of halo and hypoechogenicity were strongly predictive of BRAFV600E gene expression status in thyroid cancer. Likewise, ultrasonographic features such as hypoechogenicity, absence of halo, blurred margins, and vertical orientation were all significantly associated (P<0.05) with BRAFV600E gene expression in thyroid carcinoma, consistent with international studies’ findings (9,14,24). This may be ascribed to the invasiveness of thyroid cancer, enhanced heterogeneity of cancer cells, and alterations in the tumor microenvironment associated with the BRAFV600E gene mutation (25,26). Increased invasiveness results in tumor cells penetrating the surrounding thyroid tissue and capsule, while increased heterogeneity reflects a complex cellular composition within the tumor, contributing to indistinct margins and heterogeneous internal echoes on ultrasonography. The presence of hypoechogenicity in nodules is commonly attributed to interstitial edema, which is the buildup of fluid between follicles and within the lobules, resulting in mild to moderate hypoechogenicity. Malignant nodules are often characterized by hypoechogenic or markedly hypoechogenic appearances. A nodule with a taller-than-wide shape is usually associated with malignancy and serves as an independent risk factor. The halo observed around thyroid nodules typically stems from blood vessels encircling the periphery. Some research suggests that this halo may also result from additional pathological changes, including compressive atrophy of the thyroid tissue outside the capsule, inflammatory exudate from adjacent tissues, and interstitial edema. A thin halo may indicate benign tumors, while malignant nodules often lack a halo or present with a thick halo. However, our meta-analysis noted that the differences in solid composition, irregular shape, and microcalcifications of nodules were not statistically significant (P>0.05), discrepant from the observations of previous studies wherein microcalcifications were significantly associated with BRAFV600E mutations in thyroid cancer and considered an independent predictive factor (8,11,12,20). These discrepancies may be attributed to differences in the studied populations, with reports documenting a significant association of microcalcifications being predominantly observed in East Asian populations. Variability in interpreting ultrasound imaging features among different physicians can significantly influence these findings.

The analysis of different ultrasonographic features across the 13 included studies exposed that while the combined sensitivity of the included features was satisfactory, their specificity was relatively poor. Although this collectively enhances screening capability, it may simultaneously lead to an increase in the rate of misdiagnosis. The high sensitivity but low specificity of these predictive outcomes can be attributed to the non-specific nature of the ultrasound features and their indirect correlation with BRAFV600E mutation. Ultrasound characteristics such as internal echoes, microcalcifications, blurred margins, and the absence of a halo are notably more prevalent in nodules with BRAFV600E mutations. Nevertheless, these features might also manifest in nodules without BRAFV600E mutations, leading to reduced specificity. The BRAFV600E mutation is also associated with certain characteristics of PTC and a more aggressive tumor phenotype. Nonetheless, this association is likely indirect. Moreover, heterogeneity among the included studies was relatively high; hence, the random-effects model was adopted, and subgroup sensitivity analyses were performed using the leave-one-out approach. The significant decrease in heterogeneity observed after excluding one particular study signals that it might have been the primary contributor to the heterogeneity (12). After conducting an exhaustive review, it was determined that the current study, which exclusively focused on patients with PTC undergoing total thyroidectomy and central lymph node dissection, applied more stringent inclusion criteria compared to the remaining 12 studies, thereby accounting for the observed heterogeneity.

This study found that the prevalence of BRAFV600E mutations is lower in Western countries and higher in East Asian countries. This could be tied to genetic predispositions within the population, as well as variations in lifestyle and dietary choices. In addition, it is plausible that certain genetic traits in East Asian populations may increase their susceptibility to BRAFV600E mutations. Furthermore, disparities in iodine consumption could be linked to the frequency and variation of mutations observed in patients with thyroid cancer.

There are limitations to this study that cannot be overlooked. To begin, the patient demographic largely consisted of Asians in the included studies, with limited representation from Western populations, thereby limiting the generalizability of the findings across different ethnicities. In addition, not all studies reported data on the seven ultrasound features, and there was a lack of uniformity in their criteria across the studies. Furthermore, the exclusion of non-English language articles could have introduced inherent biases, as potentially relevant studies in other languages were not considered.

Conclusions

This meta-analysis evaluated ultrasonographic features for predicting the expression status of the BRAFV600E gene in thyroid cancer to determine their predictive value for BRAF expression status, which could mitigate the reliance on invasive diagnostic methods such as FNA biopsies and ultimately reduce the financial burden for patients. Our results collectively indicated the expression of the BRAFV600E gene in thyroid cancer was associated with hypoechoic nodules, the absence of halos, unclear margins, and a taller-than-wide shape. Notably, the absence of a halo and hypoechogenicity were identified as superior predictors of BRAFV600E gene status.

Acknowledgments

Funding: None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-24-134/rc

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-24-134/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-24-134/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Russell MD, Shonka DC Jr, Noel J, et al. Preoperative Evaluation of Thyroid Cancer: A Review of Current Best Practices. Endocr Pract 2023;29:811-21. [Crossref] [PubMed]
2. Miller KD, Nogueira L, Devasia T, et al. Cancer treatment and survivorship statistics, 2022. CA Cancer J Clin 2022;72:409-36. [Crossref] [PubMed]
3. Haugen BR, Alexander EK, Bible KC, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid 2016;26:1-133. [Crossref] [PubMed]
4. Subash A, Sinha P, Singh A. BRAF mutation and age in differentiated thyroid cancer risk stratification: Two sides of the same coin. Oral Oncol 2020;106:104732. [Crossref] [PubMed]
5. Yan C, Huang M, Li X, et al. Relationship between BRAF V600E and clinical features in papillary thyroid carcinoma. Endocr Connect 2019;8:988-96. [Crossref] [PubMed]
6. Kure S, Ishino K, Kudo M, et al. Incidence of BRAF V600E mutation in patients with papillary thyroid carcinoma: a single-institution experience. J Int Med Res 2019;47:5560-72. [Crossref] [PubMed]
7. Xu JM, Chen YJ, Dang YY, et al. Association Between Preoperative US, Elastography Features and Prognostic Factors of Papillary Thyroid Cancer With BRAF(V600E) Mutation. Front Endocrinol (Lausanne) 2019;10:902. [Crossref] [PubMed]
8. Li H, Ma J, Xi X, et al. The analysis and validation of the prediction value of conventional and contrast-enhanced ultrasonography for BRAF mutant papillary thyroid microcarcinoma. Gland Surg 2022;11:1683-96. [Crossref] [PubMed]
9. Skubisz K, Januszkiewicz-Caulier J, Cybula P, et al. Higher EU-TIRADS-Score Correlated with BRAF V600E Positivity in the Early Stage of Papillary Thyroid Carcinoma. J Clin Med 2021;10:2304. [Crossref] [PubMed]
10. Shangguan R, Hu YP, Huang J, et al. Association Between BRAF(V600E) Mutation and the American College of Radiology Thyroid Imaging, Reporting and Data System in Solitary Papillary Thyroid Carcinoma. Acad Radiol 2019;26:154-60. [Crossref] [PubMed]
11. Zhang Q, Liu BJ, Ren WW, et al. Association between BRAF V600E Mutation and Ultrasound Features in Papillary Thyroid Carcinoma Patients with and without Hashimoto's Thyroiditis. Sci Rep 2017;7:4899. [Crossref] [PubMed]
12. Wang S, Liu Z, Sun S, et al. Ultrasound features suspicious for malignancy predict the risk of BRAF mutation in papillary thyroid carcinoma. Int J Clin Exp Med 2017;10:9470-5.
13. Hahn SY, Kim TH, Ki CS, et al. Ultrasound and clinicopathological features of papillary thyroid carcinomas with BRAF and TERT promoter mutations. Oncotarget 2017;8:108946-57. [Crossref] [PubMed]
14. Kakarmath S, Heller HT, Alexander CA, et al. Clinical, Sonographic, and Pathological Characteristics of RAS-Positive Versus BRAF-Positive Thyroid Carcinoma. J Clin Endocrinol Metab 2016;101:4938-44. [Crossref] [PubMed]
15. Kwon MR, Shin JH, Park H, et al. Radiomics Study of Thyroid Ultrasound for Predicting BRAF Mutation in Papillary Thyroid Carcinoma: Preliminary Results. AJNR Am J Neuroradiol 2020;41:700-5. [Crossref] [PubMed]
16. Li Q, Yuan J, Wang Y, et al. Association between the BRAF V600E mutation and ultrasound features of the thyroid in thyroid papillary carcinoma. Oncol Lett 2017;14:1439-44. [Crossref] [PubMed]
17. Tang J, Jiang S, Ma J, et al. Nomogram based on radiomics analysis of ultrasound images can improve preoperative BRAF mutation diagnosis for papillary thyroid microcarcinoma. Front Endocrinol (Lausanne) 2022;13:915135. [Crossref] [PubMed]
18. Lee DY, Hwang SM, An JH, et al. Predicting Extrathyroidal Extension in Patients With Papillary Thyroid Microcarcinoma According to a BRAF Mutation. Clin Exp Otorhinolaryngol 2017;10:174-80. [Crossref] [PubMed]
19. Moon WJ, Choi N, Choi JW, et al. BRAF mutation analysis and sonography as adjuncts to fine-needle aspiration cytology of papillary thyroid carcinoma: their relationships and roles. AJR Am J Roentgenol 2012;198:668-74. [Crossref] [PubMed]
20. Kabaker AS, Tublin ME, Nikiforov YE, et al. Suspicious ultrasound characteristics predict BRAF V600E-positive papillary thyroid carcinoma. Thyroid 2012;22:585-9. [Crossref] [PubMed]
21. Rashid FA, Munkhdelger J, Fukuoka J, et al. Prevalence of BRAF(V600E) mutation in Asian series of papillary thyroid carcinoma-a contemporary systematic review. Gland Surg 2020;9:1878-900. [Crossref] [PubMed]
22. Bandoh N, Goto T, Kato Y, et al. BRAF V600E mutation co-existing with oncogenic mutations is associated with aggressive clinicopathologic features and poor prognosis in papillary thyroid carcinoma. Asian J Surg 2024;47:413-9. [Crossref] [PubMed]
23. Wei X, Wang X, Xiong J, et al. Risk and Prognostic Factors for BRAF(V600E) Mutations in Papillary Thyroid Carcinoma. Biomed Res Int 2022;2022:9959649. [Crossref] [PubMed]
24. Wen J, Liu H, Lin Y, et al. Correlation analysis between BRAF(V600E) mutation and ultrasonic and clinical features of papillary thyroid cancer. Heliyon 2024;10:e29955. [Crossref] [PubMed]
25. de Visser KE, Joyce JA. The evolving tumor microenvironment: From cancer initiation to metastatic outgrowth. Cancer Cell 2023;41:374-403. [Crossref] [PubMed]
26. Li DD, Zhang YF, Xu HX, et al. The role of BRAF in the pathogenesis of thyroid carcinoma. Front Biosci (Landmark Ed) 2015;20:1068-78. [Crossref] [PubMed]