Integrating US-guided FNAB, BRAF^V600E mutation, and clinicopathologic characteristics to predict cervical central lymph-node metastasis in preoperative patients with cN0 papillary thyroid carcinoma

In recent years, the incidence of thyroid cancer has substantially increased worldwide [1, 2]. The most common subtype is papillary thyroid carcinoma (PTC), accounting for almost 90% of all thyroid malignancies. Despite a rising incidence, the mortality rate from thyroid cancer has not changed significantly over the past 5 decades, which is in part related to well-developed treatment procedures [3]. Although PTC has a good prognosis, cervical central lymph-node (CLN) occult metastasis rate of patients with clinically negative central compartment lymph node (cN0) could be as high as 84.3%, which changed tumor stage and postoperative management [4]. Owing to anatomical factors including the carotid pulse and the potential for small metastases not detectable via imaging examination to be present, predicting central lymph-node metastasis (CLNM) preoperatively can be challenging. While, CLNM is generally considered a risk factor for poor prognosis in PTC, so the status of lymph node (LN) is also an important basis for judging recurrence and LN dissection [5]. At the same time, for patients with CLNM of PTC, the operation caused by persistent LNs’ recurrence will increase the risk of postoperative complications such as dyspnea, asphyxia, hypoparathyroidism, etc. [6]. Thus, it is necessary to predict CLNM in cN0 PTC patients to determine whether prophylactic central lymph-node dissection (pCLND) is needed. However, the use of pCLND in all cN0 PTC patients is still controversial [7]. Prophylactic CLND can improve disease-free survival and decrease the local recurrence rate, but it also increases the rate of surgical complication, such as recurrent laryngeal nerve injury and permanent hypoparathyroidism [6]. Therefore, routine pCLND poses additional surgical risks, and it is crucial to screen PTC patients for predictors of cervical CLNM [8].

Currently, high-resolution ultrasound (US) is recommended as the initial auxiliary approach for distinguishing between benign and malignant thyroid nodules. The combination of routine health assessments and US-guided fine-needle aspiration biopsy (FNAB) has significantly increased the rates of PTC detection and preoperative diagnosis. However, US has lower diagnostic efficiency in detecting CLNM than in detecting lateral cervical lymph node (LCLN) [9].

Several genetic mutations have been linked to the molecular pathogenesis of PTC, and examining these genes can aid in the diagnosis and prognostic evaluation of patients with PTC. The BRAF^V600E mutation is the most common genetic alteration in thyroid cancer, with an incidence ranging from approximately 30% to more than 80% [10]. This mutation has been linked to poor clinicopathological outcomes of PTC, such as LNM, disease recurrence, and patient mortality [11].

Although controversial, pCLND is still recommended in many countries [12]. Our study aimed to identify important predictors from medical information for pCLND. We retrospectively reviewed 214 consecutive PTC patients who experienced thyroidectomy plus pCLND. Univariate and multivariate analyses were also conducted to estimate the relationship between cervical CLNM and BRAF^V600E mutation status, clinical characteristics, preoperative US features, and postoperative clinicopathological features of cN0 stage PTC patients, with the goal of identifying predictors of CLNM.

Despite the favorable prognosis and high cure rate associated with most thyroid cancers, the incidence of cervical CLNM in PTC ranges from 20 to 90% according to the literature [14]. Inadequate preoperative staging by clinicians can lead to poor surgical outcomes and, in some cases, necessitate reoperation. A thorough preoperative staging evaluation can improve the selection of treatment plans and prognosis assessment, reduce the uncertainty caused by inaccurate preoperative staging, and significantly enhance the effectiveness of surgical treatment [15]. This is of great significance for improving clinical prognosis and reducing the rate of reoperation.

The preoperative diagnosis of CLNM is challenging due to anatomical factors, such as carotid pulse and potential small metastases. While preoperative ultrasound diagnosis is more sensitive for thyroid nodules, its sensitivity for preoperative diagnosis of cervical CLNM is as low as 12.1% [7]. About 60% of PTC patients have pathological evidence of cervical CLNM during their first surgery despite the absence of metastatic lesions before surgery. However, the determination of the surgical plan remains highly controversial. In 2020 [16], the American Association of Endocrine Surgeons released a new version of the “Guidelines for the Definitive Surgical Management of Thyroid Disease in Adults,” which does not recommend preventive neck lymph-node dissection for patients in T1, T2, and cN0 stages. Nevertheless, controversy remains over whether this conservative approach is appropriate for patients in China. Yang P et al. [17] believe that there is still insufficient evidence to recommend this surgical method as a routine surgical approach for patients in China. Some studies [8, 18] indicate that pCLND can prevent long-term metastasis and recurrence of thyroid cancer with high safety, even in the absence of infiltration or metastasis of lymph nodes during surgery on patients with PTC. In our study, 43.5% (93/214) of PTC patients who underwent pCLND were found to have cervical CLNM by postoperative histopathology results, suggesting that occult metastatic lymph nodes would have gone undetected if pCLND was not performed. The present study collected FNAB and BRAF molecular testing data from confirmed PTC cases and aimed to predict the occurrence of CLNM in PTC patients with cN0 through comprehensive analysis of US features and clinical indicators. The study also aimed to provide a reference for clinical decision-making regarding surgical approach.

FNAB remains a reliable diagnostic test for thyroid cancer. US-guided FNAB technology is widely used to improve diagnostic accuracy through cytologic or genomic examination [19]. Currently, the clinical basis for molecular diagnosis using FNAB is mostly derived from distinguishing between benign and malignant thyroid nodules in Class III and Class IV, with a particular focus on identifying benign nodules to prevent unnecessary surgery. Furthermore, the ability of BRAF detection in cytology is not inferior to that in pathological specimens after surgery. Multipoint and multidirectional US-guided FNAB is a new and superior puncture method compared to previous methods such as single-point and single-direction repeated puncture, and single-point and multidirectional puncture, especially for nodules less than 5 mm during percutaneous thyroid nodule biopsy (PTMC) procedures [20]. Multipoint and multidirectional US-FNAB can obtain cells from multiple locations, reducing the risk of obtaining invalid or false-negative specimens due to limited or suboptimal puncture sites, or bleeding caused by repeated puncture at a single point. This method improves the overall diagnostic yield. In our study, we used multipoint and multidirectional puncture combined with BRAFV600E mutation detection to determine the nature of thyroid nodules. This method demonstrated high diagnostic accuracy and sensitivity when compared to postoperative pathology.

This study conducted a univariate analysis to explore the relationship between cervical CLNM and clinical and US images in patients with PTC. Factors associated with cervical CLNM in PTC patients included male gender, age under 45 years, tumor size greater than 1.0 cm, nodules on both sides, contact with the thyroid capsule, presence of microcalcifications and rich blood flow, and absence of concomitant Hashimoto's thyroiditis. Initially, we used the age cutoff of 55 years based on the recent AJCC criteria[21] and found no significant correlation between age and cervical CLNM (P = 0.503). However, upon reviewing the previous version, we revised the analysis to include individuals aged 45 years, which revealed a correlation between age and cervical CLNM consistent with the previous studies [22, 23]. This study may have collected a majority of cases around the age of 43 years. Multivariate logistic regression analysis suggests that advanced age is a protective factor for the occurrence of CLNM, which is consistent with the findings of other scholars [24]. Our study found that cervical CLNM is more common in male PTC patients (P = 0.045). However, multivariate logistic regression analysis indicated that male gender is not an independent risk factor for CLNM. Other relevant studies [25] have suggested that male patients are associated with cervical CLNM, possibly due to the higher incidence of thyroid cancer in females. However, the relatively small number of male cases may have led to selection bias, and the varying inclusion and exclusion criteria in different studies indicate a need for further research and exploration of related factors.

Previous studies have demonstrated that suspicious US features, including a size greater than 1.0 cm, bilateral nodules, capsule contact, microcalcifications, and rich blood flow, as well as non-concomitant HT, are predictors of cervical CLNM in patients with PTC. These features indicate heavier tumor burden and invasiveness, and they are independent risk factors [26‐29]. For PTMC, we discovered that LNM was not associated with the nodules' size when using a maximum tumor size reference of 0.5 cm. This result is consistent with Shi et al.'s research [24]. However, further research is necessary to validate this finding. After incorporating the aforementioned risk factors into multivariate logistic analysis, bilateral nodules, microcalcifications, and non-concomitant HT were no longer significantly associated with the occurrence of cervical CLNM. Conversely, patient age less than 45 years, size greater than 1.0 cm, capsule contact, and rich blood flow were independent predictors of cervical CLNM in cN0 PTC patients. Combining US-specific changes such as tumor size, capsule contact, and rich blood flow with patient age can improve the accuracy of CLNM prediction in PTC patients, with an area under the AUC curve of 0.816.

The use of molecular diagnostic testing has had a significant impact on the diagnosis and management of thyroid nodules and thyroid cancer [30]. Our retrospective study utilized preoperative FNAB specimens to detect BRAF mutations. We discovered that 79.9% (171/214) of patients with PTC exhibited the BRAF^V600E mutation. However, we did not find significant association between BRAF^V600E mutation and CLNM. As the decision to conduct BRAF^V600E mutation tests is at the discretion of surgeons, there may be a selection bias in the study. In addition, the BRAF^V600E mutation analyses were performed in a relatively small number of patients; therefore, the rate of BRAF^V600E mutation may be underestimated, leading to the absence of a significant correlation between BRAF^V600E mutation and aggressive factors. A larger multicenter study may be necessary [24]. The BRAF^V600E mutation has been identified as the most common genetic change in PTC. The mutations activate the RAS/RAF/mitogen-activated protein kinase pathway and cause malignant cell proliferation. The BRAF^V600E mutation is more commonly found in highly aggressive subtypes, such as tall cell PTMC, indicating a significant association between BRAF^V600E mutational status and poor tumor behavior [9]. In our study, BRAF^V600E was analyzed separately, indicating that capsule contact may be a risk factor for BRAF^V600E mutation; this also confirms the correlation between BRAF^V600E mutation and high invasiveness. Lim et al. [31] analyzed the BRAF^V600E mutational status of highly aggressive PTMC and discovered that the rate of BRAF^V600E mutation was higher in patients with capsule penetration compared to those without, which is similar to our conclusion. A meta-analysis [32] involving 9924 FNAB samples confirmed that the specificity and sensitivity of BRAF^V600E mutation in diagnosing PTC were 100% and 69%, respectively. However, there is currently a debate on whether BRAF^V600E mutations are related to CLNM in PTC patients. Some scholars' research suggests that the presence of BRAF^V600E mutations can significantly increase the probability of CLNM in PTC patients [33, 34]. However, other scholars' research indicates that the BRAFV600E mutation in LNM may not be related to the invasive characteristics of LNM in PTC [7, 9, 35‐37]. Our study found that the incidence of CLNM in patients with positive BRAF^V600E mutation (77/171 = 45.03%) was higher than in patients with negative mutation (16/43 = 37.21%), suggesting that BRAF^V600E mutation may promote LNM of PTC.

Recently, predictive risk factor analysis based on clinical data has been utilized in various medical conditions. It has made significant progress. To address these concerns, this research collected and analyzed clinical data from cN0 PTC patients for retrospective analysis. Four independent risk factors were identified from numerous related factors, including age, maximum tumor size, capsule contact, and CDFI. Targeted neck lymph-node dissection was performed during surgery for precise treatment of occult CLNM.

Limitations of this study include the retrospective nature of the study, with all data collected from the same hospital, which may lead to bias. Additionally, US examination and analysis may be influenced by subjective factors, leading to errors in the judgment of cervical lymph nodes. To improve the applicability and clinical value, subsequent studies should conduct grouping studies on the patient's course of disease and conduct multicenter, large sample prospective studies.

Our analysis identified that age under 45 years, size greater than 1.0 cm, capsule contact, and rich blood flow are significant predictors for cervical CLNM in cN0 PTC patients. Prior to establishing a definitive treatment plan for the patient, a range of analyses, including US, assessments of patient clinical characteristics, and US-guided FNAB, should be performed. As not all PTC (cN0) lesions are identical, the formulation of appropriate individualized therapeutic regimens is dependent on comprehensive analyses of all of these pertinent risk factors.