Risk Stratification of Thyroid Nodules Diagnosed as Follicular Neoplasm on Core Needle Biopsy

Byeong-Joo Noh; Won Jun Kim; Jin Yub Kim; Ha Young Kim; Jong Cheol Lee; Myoung Sook Shim; Yong Jin Song; Kwang Hyun Yoon; In-Hye Jung; Hyo Sang Lee; Wooyul Paik; Dong Gyu Na

doi:10.3803/EnM.2024.2256

Articles

Page Path: HOME > Endocrinol Metab > Ahead-of print > Article

Original Article Risk Stratification of Thyroid Nodules Diagnosed as Follicular Neoplasm on Core Needle Biopsy: Byeong-Joo Noh¹, Won Jun Kim², Jin Yub Kim², Ha Young Kim², Jong Cheol Lee³, Myoung Sook Shim², Yong Jin Song³, Kwang Hyun Yoon⁴, In-Hye Jung⁵, Hyo Sang Lee⁶, Wooyul Paik^7,8, Dong Gyu Na⁷; DOI: https://doi.org/10.3803/EnM.2024.2256
Published online: May 28, 2025

¹Department of Pathology, Gangneung Asan Hospital, University of Ulsan College of Medicine, Gangneung, Korea

²Department of Endocrinology, Gangneung Asan Hospital, University of Ulsan College of Medicine, Gangneung, Korea

³Department of Otorhinolaryngology-Head and Neck Surgery, Gangneung Asan Hospital, University of Ulsan College of Medicine, Gangneung, Korea

⁴Department of Surgery, Gangneung Asan Hospital, University of Ulsan College of Medicine, Gangneung, Korea

⁵Department of Radiation Oncology, Gangneung Asan Hospital, University of Ulsan College of Medicine, Gangneung, Korea

⁶Department of Nuclear Medicine, Gangneung Asan Hospital, University of Ulsan College of Medicine, Gangneung, Korea

⁷Department of Radiolgy, Gangneung Asan Hospital, University of Ulsan College of Medicine, Gangneung, Korea

⁸Department of Radiology, Korea University Ansan Hospital, Korea University College of Medicine, Ansan, Korea

Corresponding author: Dong Gyu Na Department of Radiology, Gangneung Asan Hospital, University of Ulsan College of Medicine, 38 Bangdong-gil, Gangneung 25440, Korea Tel: +82-33-610-4310, Fax: +82-33-610-3490, E-mail: nndgna@gmail.com

• Received: November 28, 2024 • Revised: January 13, 2025 • Accepted: March 10, 2025

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

206 Views
15 Download

Full Article

Download PDF

ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
Supplementary Material
Article information
References

ABSTRACT

Background
This study assessed risk stratification and diagnostic performance for malignancy in thyroid nodules diagnosed as follicular neoplasm (FN) based on core needle biopsy (CNB) subcategories.
Methods
A total of 313 consecutive nodules (>1 cm) diagnosed as FN on CNB with corresponding surgical histology were included. FN subcategories were classified retrospectively for nodules diagnosed before 2022 (retrospective dataset) and prospectively for nodules diagnosed since 2022 (prospective dataset). CNB subcategories were determined using histologic criteria based on architectural uniformity and nuclear atypia, as modified from the 2019 Korean CNB pathology guideline. The diagnostic performance of CNB subcategories, nodule size, and ultrasound risk stratification systems (RSSs) for malignancy was assessed.
Results
CNB subcategory IVb showed a significantly higher malignancy risk compared to other subcategories in both datasets (34.5%–83.7% vs. 4.2%–13.6%, P<0.001). It was also identified as an independent predictor of malignancy in both datasets (P< 0.001), whereas nodule size and all ultrasound RSSs were not predictive of malignancy, including noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) (P≥0.079). CNB subcategory IVb demonstrated higher sensitivity for malignancy and a lower surgical rate for benign nodules compared to the nodule size criterion (>2 cm). The combined criterion of CNB subcategory IVb or nodule size >3 cm identified all malignant tumors, excluding NIFTP, in the prospective dataset.
Conclusion
CNB subcategory IVb effectively stratifies malignancy risk in thyroid nodules and outperforms nodule size (>2 cm) and ultrasound RSSs in diagnostic performance. Non-IVb nodules ≤3 cm can be safely managed with ultrasound surveillance instead of immediate surgery.
Keywords: Biopsy, large-core needle; Thyroid neoplasms; Thyroid nodule; Ultrasonography

INTRODUCTION

Ultrasound (US)-guided fine-needle aspiration (FNA) is a standard diagnostic method for evaluating thyroid nodules [1]. According to a recent meta-analysis, the follicular neoplasm (FN) category occurs in 6.1% of cases diagnosed by FNA (3.5% in Asian studies and 7.9% in Western studies), with a risk of malignancy (ROM) in surgically confirmed nodules of 28.9% (32.8% in Asian studies and 27.3% in Western studies) [2]. The estimated ROM for the FN category is approximately 30%, or 23% when noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) is excluded from malignancy, as reported by the Bethesda System for Reporting Thyroid Cytopathology [3]. In comparison, the frequency of the FN category on core needle biopsy (CNB) ranges from 8.8% to 17.5%, with an estimated ROM of 52.7%–58.1% in surgically confirmed nodules. When NIFTP is excluded, the ROM is reduced to 43.2%–44.8%, based on findings from large retrospective cohort studies [4-8]. These studies and head-to-head comparisons [9,10] consistently show that the FN category is more frequent on CNB than on FNA.

Diagnostic surgery is generally recommended for nodules classified as FN on cytology because FNA cannot reliably distinguish between benign and malignant lesions [3,11-13]. However, the optimal management strategy for FN nodules remains a topic of debate. Alternatives to immediate surgery, such as close follow-up or molecular testing, may be considered to avoid unnecessary surgical procedures in benign cases. These decisions typically depend on factors such as nodule size, cytological features, clinical presentation, and US characteristics [11-14]. Despite extensive research into malignancy risk estimation and stratification for FN nodules diagnosed on FNA, results remain controversial and inconsistent. Studies examining risk stratification based on clinical factors [15-18], nodule size [15-19], nodule growth rate [18,20,21], and US risk stratification systems (RSSs) [18,22,23] have produced conflicting findings. While the Japanese thyroid FNA cytology system incorporates cytological features into its risk stratification framework [24], this approach has not been widely adopted or integrated into other FNA systems, including the Bethesda system [3].

As CNB has proven to be more sensitive and accurate for diagnosing FN [4,5,8-10] and follicular thyroid carcinoma (FTC) [25], its use is increasing for thyroid nodules suspected of FN. Consequently, an optimized management strategy for nodules diagnosed as FN on CNB is essential to reduce unnecessary diagnostic surgeries for benign nodules. Previous studies [5,6,8] have shown that the CNB subcategory of FN with nuclear atypia is associated with a higher ROM compared to the CNB subcategory of FN without nuclear atypia. However, the risk stratification of thyroid nodules diagnosed as FN on CNB and the diagnostic performance of CNB subcategories for predicting malignancy have rarely been studied. Therefore, this study aimed to evaluate the risk stratification and diagnostic performance of CNB subcategories for malignancy in thyroid nodules diagnosed as FN on CNB.

This retrospective observational cohort study was approved by the Institutional Review Board of Gangneung Asan Hospital (IRB No. 2020-03-020), with a waiver of informed consent due to its retrospective nature. The study was conducted in accordance with the Standards for Reporting of Diagnostic Accuracy Studies statement [26].
Study design and patient characteristics: This study included consecutive patients with thyroid nodules (>1 cm) diagnosed as category IV (FN) by CNB and with final histology diagnoses confirmed by surgery between March 2017 and June 2024 at Gangneung Asan Hospital (Gangneung, Korea). During the study period, a total of 5,165 consecutive thyroid nodules underwent US-guided FNA or CNB. Of these, 4,852 nodules were excluded based on the following criteria: (1) nodules ≤1 cm (n=943); (2) nodules in which CNB was not performed (n=1,191); (3) nodules with CNB diagnoses other than FN (n=2,386); or (4) nodules without final histology diagnosis by surgery (n=332). As a result, 313 thyroid nodules (>1 cm) diagnosed as FN on CNB and with final histology diagnoses confirmed by surgery were included. These nodules were obtained from 296 patients (219 women and 77 men; mean age, 56.6±13.0 years) (Fig. 1).
Thyroid US and US-guided biopsy procedures: US examinations and US-guided biopsy procedures were performed using high-resolution color Doppler US systems equipped with linear high-frequency probes (5–14 MHz) (EPIQ7, Philips Medical Systems, Bothell, WA, USA). The US-guided CNB procedure was performed by two radiologists (Dong Gyu Na and Wooyul Paik), with 22 and 4 years of experience in thyroid imaging and intervention, respectively. The CNB procedures were performed using the freehand technique with disposable 18-gauge, single- or double-action spring-activated side-cutting needles, featuring excursion lengths of approximately 8–16 mm and specimen notch lengths of 6–15 mm (TSK Stericut or Acecut, Create Medic, Yokohama, Japan; NaBee, Flower Medical, Seongnam, Korea). The procedures were conducted under US guidance with the needle aligned parallel to the US probe, as described previously [27]. The CNB needle notch was positioned to excise a small portion (2 to 3 mm in length) of normal parenchyma at the margin of the target nodule. Strict vessel mapping along the approach route and the target nodule was performed using color Doppler US to avoid vascular injury. For procedures using a double-action CNB needle, a modified technique involving the introduction of a pre-fired stylet needle into the thyroid tissue was allowed. Biopsy sampling was performed exclusively within the target nodule in rare cases where prominent blood vessels completely encircled the nodule without an avascular window or where a large nodule replaced most of the adjacent normal thyroid parenchyma. In most cases, one or two CNB samples were obtained. At least two biopsy samplings were routinely performed for larger nodules (>3 cm) or those with heterogeneous components to minimize sampling error. The obtained tissue samples were immediately placed in biopsy cassettes and fixed in formalin to preserve tissue integrity and prevent potential loss. After withdrawing the CNB needle, manual compression was immediately applied to the biopsy site, and patients were monitored while self-compressing the site for 20 to 30 minutes post-procedure.
Categorical histology diagnosis of CNB and subcategory of follicular neoplasm: CNB histology diagnoses were classified into six categories based on the modified criteria of Korean Thyroid Association (KTA) practice guideline for thyroid CNB [28]. CNB category IV (FN) was further divided into four subcategories: IVa, FN, conventional type; IVb, FN with nuclear atypia; IVc, Oncocytic neoplasm; and IVd, FN, not otherwise specified. The detailed diagnostic criteria for each CNB subcategory of follicular-patterned thyroid nodules were primarily based on the histologic features of architectural uniformity and nuclear atypia in proliferative follicular cells (Fig. 2). Architectural uniformity refers to the consistent and uniform proliferation of follicular cells distinct from the surrounding thyroid tissue. Nuclear atypia indicates atypical nuclear features suggestive of neoplastic changes commonly associated with papillary thyroid carcinomas (PTCs) or FN. Nuclear atypia attributable to regenerative changes caused by thyroiditis was classified under categories without nuclear atypia. The presence of a nodule capsule was not considered an essential criterion for diagnosing CNB category IV in this study. These detailed histologic criteria for subcategorization of follicular-patterned thyroid nodules were developed by an expert thyroid pathologist (Byeong-Joo Noh) and have been routinely applied in clinical practice at our hospital since 2022. Previously, the diagnosis of FN on CNB was made according to the earlier thyroid CNB guidelines [29], and subcategorization of FN was not performed prior to 2022. For this study, all CNB pathology specimens of thyroid nodules preoperatively diagnosed as categories II, III, or IV between March 2017 and December 2021 and subsequently treated with thyroid surgery were retrospectively reviewed and subcategorized using the CNB diagnostic algorithm (Fig. 2) by the same pathologist (Byeong-Joo Noh) (retrospective dataset). Since 2022, CNB subcategory diagnoses of FN have been prospectively reported prior to surgery (prospective dataset).
Analysis of thyroid US images: The US features of all thyroid nodules were assessed during pre-biopsy US examinations and recorded in daily clinical practice by two radiologists (Dong Gyu Na and Wooyul Paik), as previously described [30]. The thyroid nodules were strictly classified according to the definitions provided in the US lexicons of five professional societies’ RSSs: the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TIRADS), the American Thyroid Association (ATA) system, Chinese TIRADS (C-TIRADS), European TIRADS (EU-TIRADS), and Korean TIRADS (K-TIRADS) [11,31-34]. Extrathyroidal extension status was not used for nodule classification due to the lack of standardized, specific US criteria. Nodules that could not be classified using the defined criteria were categorized as ‘intermediate suspicion’ or ‘intermediate risk’ based on the estimated malignancy risk in the ATA system and EU-TIRADS [12,35,36].
Data analysis and statistics: Continuous variables were presented as mean±standard deviation or median (interquartile range), depending on whether the data followed a parametric or nonparametric distribution. Categorical variables were reported as frequencies and percentages. The unpaired t test or Mann-Whitney U test was used to compare continuous variables, while the chi-square test or Fisher’s exact test was applied to compare categorical variables. These comparisons were made across categories of nodule size, histologic diagnoses, and malignancy risks of CNB category IV nodules according to CNB subcategories and nodule size categories, and US categories were classified as defined by each RSS. The ROM and the diagnostic performance of malignancy criteria were primarily assessed using a definition of malignancy that included NIFTP. These were also evaluated using a malignancy definition excluding NIFTP, considering its classification as a non-malignant, indolent lesion in most cases. For estimation of ROM and diagnostic performance, thyroid tumors of uncertain malignant potential were considered malignant due to their requirement for mandatory surgical intervention. Univariable and multivariable logistic regression analyses were performed to identify independent predictors of malignancy, including CNB subcategories of FN, nodule size, and US RSSs, in both retrospective and prospective datasets. Receiver operating characteristic (ROC) analysis was used to assess the diagnostic performance of these predictors. The diagnostic performance of CNB subcategories and nodule size criteria was evaluated based on sensitivity, specificity, accuracy, positive predictive value, and negative predictive value, all reported with 95% confidence intervals. Comparisons of diagnostic performance among criteria were performed on both retrospective and prospective datasets. All statistical analyses were performed using SPSS software for Windows version 28.0 (IBM Corp., Armonk, NY, USA), and MedCalc Statistical Software version 23.0.2 (MedCalc Software Ltd., Ostend, Belgium). A P value of less than 0.05 was considered statistically significant.

Demographic patient data and the characteristics of thyroid nodules: Table 1 presents the demographic data of patients and the characteristics of thyroid nodules diagnosed as category IV (FN) on CNB in the retrospective and prospective datasets. There were no significant differences in sex or age between the two datasets (P=0.593 and P=0.650, respectively). The maximum nodule diameter and the distribution of nodule sizes were also not significantly different between the two datasets (P=0.945 and P=0.788, respectively). The numbers of neoplasms, low-risk neoplasms, and malignant neoplasms were comparable between the two datasets (P≥0.092); however, the proportion of benign nodules was significantly higher in the prospective dataset compared to the retrospective dataset (64.2% vs. 51.2%, P=0.029). Additionally, the proportion of malignant tumors, including NIFTP, was significantly lower in the prospective dataset (20.8% vs. 27.5%, P=0.029). FTC or oncocytic carcinoma (OC) was the most common type of malignant tumor in the retrospective dataset (54.4%) and the overall cohort (49.4%), whereas PTC was the most prevalent malignant neoplasm in the prospective dataset (45.0%). Postoperative high-risk malignant neoplasms, as defined by the KTA guidelines [37], were identified in eight cases (3.9%) in the retrospective dataset and one case (0.9%) in the prospective dataset (P=0.143).; Neither sex nor age was significantly associated with malignancy, irrespective of the reference standard for malignancy, in the overall cohort combining both datasets (P≥0.076 and P≥0.092, respectively). In the combined cohort data from both datasets, CNB subcategory IVb was identified in all (100%) of 16 PTCs and 18 invasive encapsulated follicular variant papillary carcinomas, 53 (88.3%) of 60 NIFTPs, 20 (80%) of 25 FTCs, and five (38.5%) of 13 OCs. All high-risk malignant tumors were preoperatively diagnosed as CNB subcategory IVb in both datasets.
Risk of neoplasm and malignancy according to CNB histologic pattern of tumor capsule and surrounding thyroid tissue: The CNB category IV nodules in the prospective dataset (n=106) were classified into three histologic patterns based on the presence of a tumor capsule and surrounding thyroid tissue: pattern 1 (n=22), absence of a fibrous capsule between the tumor and surrounding thyroid tissue; pattern 2 (n=73), presence of a fibrous capsule separating the tumor from the surrounding thyroid tissue; and pattern 3 (n=11), absence of both a fibrous capsule and surrounding thyroid tissue (Supplemental Table S1). The final histologic diagnoses showed no significant differences among the three patterns in the risk of neoplasm (P=0.054), malignancy including NIFTP (P=0.372), and malignancy excluding NIFTP (P=0.357).
Risk of malignancy of CNB category IV nodules according to CNB subcategory, nodule size, and US category of RSS: Tables 2, 3 present the ROMs for category IV nodules based on the CNB subcategory, nodule size, and US category of RSS in the retrospective and prospective datasets. CNB subcategory IVb nodules exhibited significantly higher ROMs compared to other subcategories in both datasets, irrespective of the reference standard for malignancy (34.5%–83.7% vs. 4.2%–13.6%, P<0.001, respectively). However, ROMs showed no significant differences based on nodule size criteria (P≥0.237) or US categories of K-TIRADS, ATA, and ACR TIRADS in either dataset, regardless of the reference standard for malignancy (P≥0.114) (Supplemental Tables S2, S3). In C-TIRADS, the ROMs for US categories, including C-TR5 or C-TR4 were significantly higher than for other categories in the retrospective dataset using malignancy excluding NIFTP as the reference standard (64.3% vs. 24.9%, P=0.001) and in the prospective dataset irrespective of the malignancy definition (66.7%–83.3% vs. 18.0%–33.0%, P≤0.022). In EU-TIRADS, the ROM for the EU-TIRADS 4 category was significantly higher than for other categories in the prospective dataset when using malignancy, including NIFTP as the reference standard (47.6% vs. 28.1%, P=0.001).
Association of CNB subcategory, nodule size, and US RSS with malignancy in CNB category IV nodules: Tables 2, 3 present the results of univariable and multivariable logistic regression analyses for the CNB subcategory, nodule size, and US categories of K-TIRADS in predicting malignancy. Multivariable analysis revealed that CNB subcategory IVb was an independent predictor of malignancy in both datasets, regardless of the reference standard for malignancy. In the retrospective dataset, the odds ratios (ORs) for malignancy including NIFTP were 34.9–41.1 (P<0.001), and for malignancy excluding NIFTP, the ORs were 6.0–7.1 (P<0.001). In the prospective dataset, the ORs were 8.5–11.7 for malignancy including NIFTP (P<0.001), and 6.0–15.2 for malignancy excluding NIFTP (P≤0.001) (Supplemental Tables S2, S3). However, nodule size and all US RSSs were not independently predictive of malignancy including NIFTP, in either dataset (P≥0.079). The C-TIRADS category of C-TR5 or C-TR4 was an independent predictor of malignancy only in the retrospective dataset when using a reference standard excluding NIFTP (OR, 7.5; P=0.013). In contrast, none of the other US RSSs were independently predictive of malignancy excluding NIFTP in either dataset.
Diagnostic performance of CNB subcategory, nodule size, and US RSS for malignancy in CNB category IV nodules: The area under the ROC curve for the CNB subcategory using a cutoff of IVb was significantly greater than that for nodule size and all US RSSs in the retrospective dataset (0.706–0.853 vs. 0.503–0.564, P≤0.011) and the prospective dataset (0.732–0.757 vs. 0.500–0.527, P≤0.021) regardless of the reference standard for malignancy. Nodule size and US RSSs showed no discriminative power between benign and malignant nodules in the ROC analysis for the retrospective dataset (P≥0.632 and P≥0.135, respectively) and the prospective dataset (P≥0.687 and P≥0.633, respectively), irrespective of the malignancy reference standard.; Table 4 summarizes the diagnostic performance of CNB subcategory IVb and nodule size criteria (with 2 and 3 cm cutoffs) for diagnosing malignancy in both retrospective and prospective datasets. The sensitivities of the CNB subcategory IVb criterion were significantly higher (80.7%–86.1% vs. 61.4%, P≤0.043), and the surgical rates for benign nodules were significantly lower (16.0%–38.7% vs. 62.7%–63.2%, P≤0.001), compared to the nodule size criterion with a 2 cm cutoff in the retrospective dataset, regardless of the malignancy reference standard. In the prospective dataset, the sensitivities of the CNB subcategory IVb criterion were higher than those of the 2 cm nodule size criterion (84.2%–90.9% vs. 60.5%–63.6%, P≥0.064), though the difference was not statistically significant. However, the surgical rates for benign nodules were significantly lower (38.2%–45.2% vs. 66.7%–69.2%, P≤0.011) for the CNB subcategory IVb criterion, irrespective of the reference standard for malignancy. The combined criterion of CNB subcategory IVb or a nodule size larger than 3 cm demonstrated significantly higher sensitivity (89.1% vs. 61.4%, P<0.001) and a lower surgical rate for benign nodules (45.3% vs. 63.2%, P=0.001) compared to the 2 cm nodule size criterion in the retrospective dataset, when using a malignancy reference standard that includes NIFTP. In the prospective dataset, this combined criterion also showed significantly higher sensitivity (94.7% vs. 60.5%, P<0.001) and a comparable surgical rate for benign nodules (60.3% vs. 69.1%, P=0.286). Notably, in the prospective dataset, all malignant tumors excluding NIFTP were correctly identified using the combined criterion of CNB subcategory IVb or a nodule size larger than 3 cm (Table 4).

DISCUSSION

Our study demonstrated that CNB subcategory IVb was an independent predictor of malignancy in both the retrospective and prospective datasets, regardless of the reference standard for malignancy. In contrast, nodule size and all US RSSs were not predictive of malignancy including NIFTP, and lacked diagnostic value for malignancy in either dataset. The CNB subcategory IVb demonstrated superior diagnostic performance, offering higher sensitivity for malignancy and a lower surgical rate for benign nodules compared to the nodule size criterion (>2 cm) in both datasets. The combined criterion of CNB subcategory IVb or a nodule size larger than 3 cm showed very high sensitivity, detecting all malignant tumors excluding NIFTP, while maintaining a surgical rate for benign nodules comparable to the nodule size criterion (>2 cm) in the prospective dataset.

The CNB category of follicular-patterned thyroid nodules was primarily determined by the architectural uniformity and nuclear atypia of proliferative follicular cells. Architectural uniformity, distinct from the adjacent thyroid tissue, differentiated category IV from category III or category II, depending on the proportion of uniformity, while nuclear atypia was used to further subcategorize categories III and IV (Fig. 2). The presence of a nodule capsule was not considered an essential criterion for diagnosing FN on CNB in this study. However, when distinguishing between category IV and category III based on architectural uniformity in the CNB specimen is challenging, the presence of a nodule capsule increases the likelihood of category IV. This approach differs slightly from the general concept in current CNB practice guidelines, which typically classify follicular proliferative lesions as FN only when a tumor capsule is present [28,38]. Our histologic approach to the follicular capsule in diagnosing FN was based on several considerations: (1) the absence of a capsule in CNB specimens does not specifically indicate nonneoplastic or benign changes. Follicular-patterned malignant tumors, such as NIFTP or widely invasive FTC, may exhibit partial encapsulation or even lack a capsule. Similarly, partial encapsulation can sometimes occur in thyroid follicular nodular disease (nodular hyperplasia); (2) the malignancy rate did not significantly differ based on the presence or absence of an apparent tumor capsule in CNB specimens [39]. This finding is consistent with our prospective cohort data, which showed no significant difference in malignancy rates regardless of tumor capsule presence; (3) in certain cases, such as large nodules replacing most of the normal thyroid parenchyma or markedly hypervascular nodules with vessels fully encasing the tumor margin, accurate sampling of the tumor capsule may not be technically feasible; (4) during processing, CNB specimens manipulated for formalin-fixed paraffin embedding may become disoriented, making it difficult to definitively determine the presence of a capsule. Therefore, unlike the surgical diagnosis of FN, the identification of a nodule capsule on a CNB specimen may not serve as a mandatory diagnostic criterion for FN. The CNB category IV nodules without an identified nodule capsule classified in our study may be categorized as CNB category IIIb (indeterminate follicular lesion with architectural atypia), IIId (indeterminate follicular lesion with Hürthle cell changes), or IIIc (indeterminate follicular lesion with nuclear and architectural atypia) based on the presence of nuclear atypia, according to the CNB reporting system guideline proposed by the KTA [28]. Our CNB criteria may reduce the proportion of indeterminate lesions and increase the proportion of FN diagnoses compared to the KTA CNB criteria. The CNB subcategory IVb, characterized by FN with nuclear atypia, showed a significantly higher ROM compared to other FN subcategories without nuclear atypia. This finding aligns with previous cytology studies, which demonstrated that the presence of nuclear atypia in FNA category IV nodules increases the ROM [40-42].

The relatively lower ROM of the FN category on CNB in our study population compared to previous reports [4-8] could primarily be attributed to differences in disease prevalence and spectrum [43], as well as potential variations in the histologic CNB criteria for diagnosing FN. The prevalence and severity of malignant tumors, including follicular-patterned malignancies, may be lower in our community hospital study population compared to the populations of nationwide large tertiary hospitals where previous studies were conducted.

Our results provide several important insights into the management of CNB category IV nodules. First, the combined criterion of CNB subcategory IVb or a nodule size greater than 3 cm can effectively triage CNB category IV nodules for immediate surgery. This criterion detects the majority of malignant tumors, including NIFTP, all large (>3 cm) malignant tumors, and all high-risk malignant tumors. Second, CNB subcategory non-IVb nodules with a size ≤3 cm can be conservatively managed with US surveillance rather than immediate surgery, provided there is no suspicion of gross tumor invasion of adjacent organs or metastasis. This approach could help avoid unnecessary surgery in approximately 40%–50% of benign nodules. Third, nodule size and US RSSs are not reliable tools for stratifying malignancy risk in nodules diagnosed as FN on CNB. However, larger nodule sizes (>3 or 4 cm) should still be considered as potential poor prognostic factors when managing these nodules [44,45]. Additionally, the development of an effective US RSS specifically designed for nodules diagnosed as FN on biopsy is necessary in the future.

This study has several limitations. First, the exclusion of patients who did not undergo surgery may have introduced selection bias, potentially leading to an overestimation of the malignancy risk. Second, while our findings were validated in a prospective cohort at a single institution, external validation is necessary to confirm the generalizability of these results. Third, the estimation of nodule growth rate was not conducted in this retrospective study because only a minority proportion of patients underwent US follow-up for ≥1 year prior to delayed surgery, and the timing of these follow-ups was heterogeneous.

In conclusion, the malignancy risk of CNB category IV nodules was effectively stratified by the CNB subcategory IVb, which demonstrated superior diagnostic performance compared to nodule size or US RSSs. CNB subcategory non-IVb nodules with a size ≤3 cm can be managed conservatively with US surveillance, thereby avoiding unnecessary immediate surgery.

Supplementary Material

Supplemental Table S1.

Postoperative Diagnosis according to the Histologic Patterns of Tumor Capsule and Surrounding Thyroid Tissue on CNB Specimens in Prospective Dataset

enm-2024-2256-Supplemental-Table-S1.pdf

Supplemental Table S2.

Risk of Malignancy and Association of CNB Subcategories, Nodule Size, US Categories with Malignancy in Retrospective Dataset (n=207)

enm-2024-2256-Supplemental-Table-S2.pdf

Supplemental Table S3.

Risk of Malignancy and Association of CNB Subcategories, Nodule Size, US Categories with Malignancy in Prospective Dataset (n=106)

enm-2024-2256-Supplemental-Table-S3.pdf

Article information

CONFLICTS OF INTEREST

No potential conflict of interest relevant to this article was reported.

ACKNOWLEDGMENTS

This research was supported by the Medical Research Promotion Program through the Gangneung Asan Hospital, funded by the Asan Foundation (2020IC001).

AUTHOR CONTRIBUTIONS

Conception or design: B.J.N., D.G.N. Acquisition, analysis, or interpretation of data: B.J.N., J.C.L., Y.J.S., K.H.Y., W.P., D. G.N. Drafting the work or revising: B.J.N., W.J.K., J.Y.K., H.Y.K., M.S.S., I.H.J., H.S.L., D.G.N. Final approval of the manuscript: B.J.N., W.J.K., J.Y.K., H.Y.K., J.C.L., M.S.S., Y. J.S., K.H.Y., I.H.J., H.S.L., W.P., D.G.N.

Fig. 1.

Flow diagram of patient enrollment. US, ultrasound; FNA, fine-needle aspiration; CNB, core needle biopsy; FN, follicular neoplasm.

Fig. 2.

Thyroid core needle biopsy diagnostic algorithm for follicular-patterned nodules. Category IIIa is assigned only when the nuclear atypia is evident. If the nodule shows uniformity with equivocal nuclear atypia less than 10%, its category is determined by the criteria of architectural uniformity without nuclear atypia. IIIa, indeterminate follicular lesion with nuclear atypia; IIIb, indeterminate follicular lesion with architectural atypia; IIIc, indeterminate follicular lesion with nuclear and architectural atypia; IIId, indeterminate follicular lesion with oncocytic cell changes; IVa, follicular neoplasm, conventional type; IVb, follicular neoplasm with nuclear atypia; IVc, oncocytic neoplasm.

Table 1.

Demographic Patient Data and the Characteristics of Thyroid Nodules in Retrospective and Prospective Datasets

Characteristic	Retrospective dataset	Prospective dataset	P value
No. of patients	195	101
No. of female patients	146 (74.9)	73 (71.9)	0.593
Age, yr	56.9±13.2	56.1±12.8	0.650
No. of nodules	207	106
Maximal nodule diameter, cm^a	2.4 (1.6–3.6)	2.5 (1.7–3.6)	0.945
1–2	78 (37.7)	36 (34.0)	0.788
2.1–3	58 (28.0)	34 (32.1)
3.1–4	29 (14.0)	17 (16.0)
>4	42 (20.3)	19 (17.9)
No. of neoplasm	200 (96.6)	100 (94.3)	0.376
No. of malignant tumor including NIFTP	101 (48.8)	38 (35.8)	0.029
No. of malignant tumor excluding NIFTP	57 (27.5)	22 (20.8)	0.191
Histologic diagnosis by surgery
Benign nodules	106 (51.2)	68 (64.2)	0.029
Thyroid follicular nodular disease	6 (5.7)	6 (8.8)
FA/OA	98 (92.5)	62 (91.2)
Other benign lesions	2 (1.9)	0
Low-risk neoplasm	44 (21.3)	18 (17.0)	0.369
NIFTP	44 (100.0)	16 (88.9)
Thyroid tumors of uncertain malignant potential	0	2 (11.1)
Malignant neoplasm	57(27.5)	20 (18.9)	0.092
PTC	7 (12.3)	9 (45.0)
IE-FVPTC	15 (26.3)	3 (15.0)
FTC/OC	31 (54.4)	7 (35.0)
Other malignant tumors	4 (7.0)	1 (5.0)
High-risk malignant tumors	8 (3.9)	1 (0.9)	0.143
PDTC	4	1
Widely invasive FTC/OC	4^b	0
FTC with vascular invasion (≥4)	1^b	0
Postoperative T4 stage (pT4)	1^b	0
Metastatic cervical lymph node (>3 cm)	0	0
Distant metastasis	0	0

Values are expressed as number (%), mean±standard deviation, or median (interquartile range).

NIFTP, noninvasive follicular thyroid neoplasm with papillary-like nuclear feature; FA, follicular adenoma; OA, oncocytic adenoma; PTC, papillary thyroid carcinoma; IE-FVPTC, Invasive encapsulated follicular variant papillary carcinoma; FTC, follicular thyroid carcinoma; OC, oncocytic carcinoma; PDTC, poorly differentiated thyroid carcinoma.

^a Measured on ultrasound;

^b Same one patient overlaps.

Table 2.

Risk of Malignancy and Association of Core Needle Biopsy Subcategories, Nodule Size, K-TIRADS Categories with Malignancy in Retrospective Dataset (n=207)

Variable	ROM with NIFTP, %	Univariable analysis		Multivariable analysis		ROM without NIFTP, %	Univariable analysis		Multivariable analysis
Variable	ROM with NIFTP, %	OR (95% CI)	P value	OR (95% CI)	P value	ROM without NIFTP, %	OR (95% CI)	P value	OR (95% CI)	P value
CNB subcategory
IVa	14.8 (4/27)					7.4 (2/27)
IVb	83.7 (87/104)	32.5 (15.1–70.0)	<0.001	39.1 (16.9–90.5)	<0.001	44.2 (46/104)	6.6 (3.2–13.8)	<0.001	6.3 (3.0–13.4)	<0.001
IVc	13.7 (10/73)					12.3 (9/73)
IVd	0.0 (0/3)					0.0 (0/3)
IVa, IVc, IVd	13.6 (14/103)	1.0		1.0		10.7 (11/103)	1.0		1.0
Size, cm
>4.0	57.1 (24/42)	1.3 (0.6–2.8)	0.455	0.9 (0.3–2.8)	0.806	38.1 (16/42)	1.6 (0.7–3.5)	0.268	1.5 (0.6–4.0)	0.370
3.1–4.0	34.5 (10/29)	0.5 (0.2–1.3)	0.155	0.4 (0.1–1.5)	0.164	17.2 (5/29)	0.5 (0.2–1.6)	0.251	0.6 (0.2–2.1)	0.445
2.1–3.0	48.3 (28/58)	0.9 (0.5–1.8)	0.842	1.1 (0.4–3.0)	0.887	24.1 (14/58)	0.8 (0.4–1.8)	0.595	0.9 (0.4–2.3)	0.894
≤2.0	50.0 (39/78)	1.0		1.0		28.2 (22/78)	1.0		1.0
K-TIRADS
5 (high suspicion)	54.8 (17/31)	1.4 (0.6–3.4)	0.394	0.4 (0.1–1.5)	0.172	35.5 (11/31)	1.7 (0.7–4.1)	0.285	1.2 (0.4–3.6)	0.802
4 (intermediate suspicion)	49.1 (53/108)	1.2 (0.6–2.1)	0.652	0.7 (0.3–1.7)	0.417	26.9 (29/108)	1.1 (0.6–2.2)	0.785	0.9 (0.4–2.1)	0.890
2 (benign) or 3 (low suspicion)	45.6 (31/68)	1.0		1.0		25.0 (17/68)	1.0		1.0

Data in parentheses are raw data.

K-TIRADS, Korean Thyroid Imaging Reporting and Data System; ROM, risk of malignancy; NIFTP, noninvasive follicular thyroid neoplasm with papillary-like nuclear feature; OR, odds ratio; CI, confidence interval; CNB, core needle biopsy.

Table 3.

Risk of Malignancy and Association of Core Needle Biopsy Subcategories, Nodule Size, K-TIRADS Categories with Malignancy in Prospective Dataset (n=106)

Variable	ROM with NIFTP, %	Univariable analysis		Multivariable analysis		ROM without NIFTP, %	Univariable analysis		Multivariable analysis
Variable	ROM with NIFTP, %	OR (95% CI)	P value	OR (95% CI)	P value	ROM without NIFTP, %	OR (95% CI)	P value	OR (95% CI)	P value
CNB subcategory
IVa	29.4 (5/17)					5.9 (1/17)
IVb	55.2 (32/58)	8.6 (3.2–23.4)	<.001	11.7 (3.9–35.2)	<.001	34.5 (20/58)	12.1 (2.7–55.1)	<.001	14.5 (3.0–69.9)	<0.001
IVc	3.3 (1/30)					3.3 (1/30)
IVd	0.0 (0/1)					0.0 (0/1)
IVa, IVc, IVd	12.5 (6/48)	1.0		1.0		4.2 (2/48)	1.0		1.0
Size, cm
>4.0	31.6 (6/19)	0.6 (0.2–2.1)	0.465	0.4 (0.1–2.2)	0.306	26.3 (5/19)	1.3 (0.3–4.5)	0.734	1.0 (0.2–5.8)	0.999
3.1–4.0	47.1 (8/17)	1.2 (0.4–4.0)	0.712	1.3 (0.3–5.8)	0.694	17.6 (3/17)	0.8 (0.2–3.3)	0.702	0.7 (0.1–3.9)	0.673
2.1–3.0	26.5 (9/34)	0.5 (0.2–1.4)	0.184	0.4 (0.1–1.2)	0.099	17.6 (6/34)	0.8 (0.2–2.4)	0.633	0.6 (0.2–2.4)	0.472
≤2.0	41.7 (15/36)	1.0		1.0		22.2 (8/36)	1.0		1.0
K-TIRADS
5 (high suspicion)	36.8 (7/19)	1.0 (0.3–3.3)	0.972	0.3 (0.1–1.5)	0.252	26.3 (5/19)	1.1 (0.3–4.1)	0.868	0.5 (0.1–2.5)	0.373
4 (intermediate suspicion)	35.2 (19/54)	1.0 (0.4–2.3)	0.911	0.5 (0.1–1.7)	0.142	16.7 (9/54)	0.6 (0.2–1.8)	0.39	0.4 (0.1–1.7)	0.211
2 (benign) or 3 (low suspicion)	36.4 (12/33)	1.0		1.0		24.2 (8/33)	1.0		1.0

Data in parentheses are raw data.

Table 4.

Diagnostic Performance of Core Needle Biopsy Subcategory (IVb), Nodule Size, and Combined Criteria for the Diagnosis of Malignancy

	Percentage (no./total no) (95% confidence interval)
	Sensitivity	Specificity	PPV	NPV	Accuracy	Surgical rate of benign nodules^a
Retrospective dataset (n=207)
Diagnostic criteria for malignancy with NIFTP
Subcategory IVb	86.1 (87/101)	84.0 (89/106)	83.7 (87/104)	86.4 (89/103)	85.0 (176/207)	16.0 (17/106)
Subcategory IVb	(79.4–92.9)	(77.0–91.0)	(76.6–90.8)	(79.8–93.0)	(80.2–89.9)	(9.1–23.0)
Nodule size >2 cm	61.4 (62/101)	36.8 (39/106)	48.1 (62/129)	50.0 (39/78)	48.8 (101/207)	63.2 (67/106)
Nodule size >2 cm	(51.9–70.9)	(27.6–46.0)	(39.4–56.7)	(38.9–61.1)	(42.0–55.6)	(54.0–72.4)
Nodule size >3 cm	33.7 (34/101)	65.1 (69/106)	47.9 (34/71)	50.7 (69/136)	49.8 (103/207)	34.9 (37/106)
Nodule size >3 cm	(24.5–42.9)	(56.0–74.2)	(36.3–59.5)	(42.3–59.1)	(43.0–56.6)	(25.8–44.0)
Subcategory IVb or size >3 cm	89.1 (90/101)	54.7 (58/106)	65.2 (90/138)	84.1 (58/69)	71.5 (148/207)	45.3 (48/106)
Subcategory IVb or size >3 cm	(83.0–95.2)	(45.2–64.2)	(57.3–73.2)	(75.4–92.7)	(65.4–77.7)	(35.8–54.8)
Diagnostic criteria for malignancy without NIFTP
Subcategory IVb	80.7 (46/57)	61.3 (92/150)	44.2 (46/104)	89.3 (92/103)	66.7 (138/207)	38.7 (58/150)
Subcategory IVb	(70.5–91.0)	(53.5–69.1)	(34.7–53.8)	(83.4–95.3)	(60.2–73.1)	(30.9–46.5)
Nodule size >2 cm	61.4 (35/57)	37.3 (56/150)	27.1 (35/129)	71.8 (56/78)	44.0 (91/207)	62.7 (94/150)
Nodule size >2 cm	(48.8–74.0)	(29.6–45.1)	(19.5–34.9)	(61.8–81.8)	(37.2–50.7)	(54.9–70.4)
Nodule size >3 cm	36.8 (21/57)	66.7 (100/150)	29.6 (21/71)	73.5 (100/136)	58.5 (121/207)	33.3 (50/150)
Nodule size >3 cm	(24.3–49.4)	(59.1–74.2)	(19.0–40.2)	(66.1–80.9)	(51.7–65.2)	(25.8–40.9)
Subcategory IVb or size >3 cm	84.2 (48/57)	40.0 (60/150)	34.8 (48/138)	87.0 (60/69)	52.2 (108/207)	60.0 (90/150)
Subcategory IVb or size >3 cm	(74.7–93.7)	(32.2–47.8)	(26.8–42.7)	(79.0–94.9)	(45.4–59.0)	(52.2–67.8)
Prospective dataset (n=106)
Diagnostic criteria for malignancy with NIFTP
Subcategory IVb	84.2 (32/38)	61.8 (42/68)	55.2 (32/58)	87.5 (42/48)	69.8 (74/106)	38.2 (26/68)
Subcategory IVb	(72.6,95.8)	(50.2–73.3)	(42.4–68.0)	(78.1–96.9)	(61.1–78.6)	(26.7–49.8)
Nodule size >2 cm	60.5 (23/38)	30.9 (21/68)	32.9 (23/70)	58.3 (21/36)	41.5 (44/106)	69.1 (47/68)
Nodule size >2 cm	(45.0–76.1)	(19.9–41.9)	(21.9–43.9)	(42.2–74.4)	(32.1–50.9)	(58.1–80.1)
Nodule size >3 cm	36.8 (14/38)	67.6 (46/68)	38.9 (14/36)	65.7 (46/70)	56.6 (60/106)	32.4 (22/68)
Nodule size >3 cm	(21.5–52.2)	(56.5–78.8)	(23.0–54.8)	(54.6–76.8)	(47.2–66.0)	(21.2–43.5)
Subcategory IVb or size >3 cm	94.7 (36/38)	39.7 (27/68)	46.8 (36/77)	93.1 (27/29)	59.4 (63/106)	60.3 (41/68)
Subcategory IVb or size >3 cm	(87.6–100.0)	(28.1–51.3)	(35.6–57.9)	(83.9–100.0)	(50.1–68.8)	(48.7–71.9)
Diagnostic criteria for malignancy without NIFTP
Subcategory IVb	90.9 (20/22)	54.8 (46/84)	34.5 (20/58)	95.8 (46/48)	62.3 (66/106)	45.2 (38/84)
Subcategory IVb	(78.9–100.0)	(44.1–65.4)	(22.3–46.7)	(90.2–100.0)	(53.0–71.5)	(34.6–55.9)
Nodule size >2 cm	63.6 (14/22)	33.3 (28/84)	20.0 (14/70)	77.8 (28/36)	39.6 (42/106)	66.7 (56/84)
Nodule size >2 cm	(43.5–83.7)	(23.3–43.4)	(10.6–29.4)	(64.2–91.4)	(30.3–48.9)	(56.6–76.8)
Nodule size >3 cm	36.4 (8/22)	66.7 (56/84)	22.2 (8/36)	80.0 (56/70)	60.4 (64/106)	33.3 (28/84)
Nodule size >3 cm	(16.3–56.5)	(56.6–76.8)	(8.6–35.8)	(70.6–89.4)	(51.1–69.7)	(23.3–43.4)
Subcategory IVb or size >3 cm	100.0 (22/22)	34.5 (29/84)	28.6 (22/77)	100.0 (29/29)	48.1 (51/106)	65.5 (55/84)
Subcategory IVb or size >3 cm	(100.0–100.0)	(24.4–44.7)	(18.5–38.7)	(100.0–100.0)	(38.6–57.6)	(55.3,75.6)

Data in parentheses are raw data.

PPV, positive predictive value; NPV, negative predictive value; NIFTP, noninvasive follicular thyroid neoplasm with papillary-like nuclear feature.

^a Includes NIFTP in the case of diagnostic criteria for malignancy without NIFTP.

References

1. Ha EJ, Lim HK, Yoon JH, Baek JH, Do KH, Choi M, et al. Primary imaging test and appropriate biopsy methods for thyroid nodules: guidelines by Korean Society of Radiology and National Evidence-Based Healthcare Collaborating Agency. Korean J Radiol 2018;19:623–31.PubMed PMC
2. Vuong HG, Ngo HT, Bychkov A, Jung CK, Vu TH, Lu KB, et al. Differences in surgical resection rate and risk of malignancy in thyroid cytopathology practice between Western and Asian countries: a systematic review and meta-analysis. Cancer Cytopathol 2020;128:238–49.PubMed
3. Ali SZ, Baloch ZW, Cochand-Priollet B, Schmitt FC, Vielh P, VanderLaan PA. The 2023 Bethesda System for Reporting Thyroid Cytopathology. Thyroid 2023;33:1039–44.Article PubMed
4. Suh CH, Baek JH, Choi YJ, Kim TY, Sung TY, Song DE, et al. Efficacy and safety of core-needle biopsy in initially detected thyroid nodules via propensity score analysis. Sci Rep 2017;7:8242.Article PubMed PMC PDF
5. Chung SR, Baek JH, Lee JH, Lee YM, Sung TY, Chung KW, et al. Risk of malignancy according to the sub-classification of atypia of undetermined significance and suspicious follicular neoplasm categories in thyroid core needle biopsies. Endocr Pathol 2019;30:146–54.PubMed
6. Na HY, Woo JW, Moon JH, Choi JY, Jeong WJ, Kim YK, et al. Preoperative diagnostic categories of noninvasive follicular thyroid neoplasm with papillary-like nuclear features in thyroid core needle biopsy and its impact on risk of malignancy. Endocr Pathol 2019;30:329–39.Article PubMed PDF
7. Son HM, Kim JH, Kim SC, Yoo RE, Bae JM, Seo H, et al. Distribution and malignancy risk of six categories of the pathology reporting system for thyroid core-needle biopsy in 1,216 consecutive thyroid nodules. Ultrasonography 2020;39:159–65.Article PubMed PMC PDF
8. Kim K, Bae JS, Kim JS, Jung SL, Jung CK. Diagnostic performance of thyroid core needle biopsy using the revised reporting system: comparison with fine needle aspiration cytology. Endocrinol Metab (Seoul) 2022;37:159–69.Article PubMed PMC PDF
9. Hong MJ, Na DG, Kim SJ, Kim DS. Role of core needle biopsy as a first-line diagnostic tool for thyroid nodules: a retrospective cohort study. Ultrasonography 2018;37:244–53.Article PubMed PDF
10. Joo L, Na DG, Kim JH, Seo H. Comparison of core needle biopsy and repeat fine-needle aspiration in avoiding diagnostic surgery for thyroid nodules initially diagnosed as atypia/follicular lesion of undetermined significance. Korean J Radiol 2022;23:280–8.Article PubMed PMC PDF
11. Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, 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.PubMed PMC
12. Durante C, Hegedus L, Czarniecka A, Paschke R, Russ G, Schmitt F, et al. 2023 European Thyroid Association clinical practice guidelines for thyroid nodule management. Eur Thyroid J 2023;12:e230067.Article PubMed PMC
13. Park YJ, Lee EK, Song YS, Kang SH, Koo BS, Kim SW, et al. Korean Thyroid Association management guidelines for patients with thyroid nodules 2024. Int J Thyroidol 2024;17:208–44.
14. Hirokawa M, Auger M, Jung CK, Callegari FM. Thyroid FNA cytology: the Eastern versus Western perspectives. Cancer Cytopathol 2023;131:415–20.Article PubMed
15. Schlinkert RT, van Heerden JA, Goellner JR, Gharib H, Smith SL, Rosales RF, et al. Factors that predict malignant thyroid lesions when fine-needle aspiration is “suspicious for follicular neoplasm”. Mayo Clin Proc 1997;72:913–6.PubMed
16. Raber W, Kaserer K, Niederle B, Vierhapper H. Risk factors for malignancy of thyroid nodules initially identified as follicular neoplasia by fine-needle aspiration: results of a prospective study of one hundred twenty patients. Thyroid 2000;10:709–12.PubMed
17. Gulcelik NE, Gulcelik MA, Kuru B. Risk of malignancy in patients with follicular neoplasm: predictive value of clinical and ultrasonographic features. Arch Otolaryngol Head Neck Surg 2008;134:1312–5.PubMed
18. Hirokawa M, Suzuki A, Kawakami M, Kudo T, Miyauchi A. Criteria for follow-up of thyroid nodules diagnosed as follicular neoplasm without molecular testing: the experience of a high-volume thyroid centre in Japan. Diagn Cytopathol 2022;50:223–9.Article PubMed PMC PDF
19. Cho YY, Ahn SH, Lee EK, Park YJ, Choi D, Kim BY, et al. Malignancy risk of follicular neoplasm (Bethesda IV) with variable cutoffs of tumor size: a systemic review and meta-analysis. J Clin Endocrinol Metab 2024;109:1383–92.PubMed
20. Nakamura H, Hirokawa M, Ota H, Kihara M, Miya A, Miyauchi A. Is an increase in thyroid nodule volume a risk factor for malignancy? Thyroid 2015;25:804–11.PubMed
21. Kim M, Chung SR, Jeon MJ, Han M, Lee JH, Song DE, et al. Determining whether tumor volume doubling time and growth rate can predict malignancy after delayed diagnostic surgery of follicular neoplasm. Thyroid 2019;29:1418–24.Article PubMed
22. Hong MJ, Na DG, Baek JH, Sung JY, Kim JH. Cytology-ultrasonography risk-stratification scoring system based on fine-needle aspiration cytology and the Korean-Thyroid Imaging Reporting and Data System. Thyroid 2017;27:953–9.Article PubMed
23. Slowinska-Klencka D, Wysocka-Konieczna K, Klencki M, Popowicz B. Diagnostic value of six thyroid imaging reporting and data systems (TIRADS) in cytologically equivocal thyroid nodules. J Clin Med 2020;9:2281.PubMed PMC
24. Satoh S, Yamashita H, Kakudo K. Thyroid cytology: the Japanese System and experience at Yamashita Thyroid Hospital. J Pathol Transl Med 2017;51:548–54.Article PubMed PDF
25. Park KW, Shin JH, Hahn SY, Oh YL, Kim SW, Kim TH, et al. Ultrasound-guided fine-needle aspiration or core needle biopsy for diagnosing follicular thyroid carcinoma? Clin Endocrinol (Oxf) 2020;92:468–74.Article PubMed PDF
26. Cohen JF, Korevaar DA, Altman DG, Bruns DE, Gatsonis CA, Hooft L, et al. STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration. BMJ Open 2016;6:e012799.Article PubMed PMC
27. Hong MJ, Na DG, Lee H. Diagnostic efficacy and safety of core needle biopsy as a first-line diagnostic method for thyroid nodules: a prospective cohort study. Thyroid 2020;30:1141–9.Article PubMed
28. Jung CK, Baek JH, Na DG, Oh YL, Yi KH, Kang HC. 2019 Practice guidelines for thyroid core needle biopsy: a report of the Clinical Practice Guidelines Development Committee of the Korean Thyroid Association. J Pathol Transl Med 2020;54:64–86.Article PubMed PMC PDF
29. Jung CK, Min HS, Park HJ, Song DE, Kim JH, Park SY, et al. Pathology reporting of thyroid core needle biopsy: a proposal of the Korean Endocrine Pathology Thyroid Core Needle Biopsy Study Group. J Pathol Transl Med 2015;49:288–99.PubMed PMC
30. Na DG, Paik W, Cha J, Gwon HY, Kim SY, Yoo RE. Diagnostic performance of the modified Korean Thyroid Imaging Reporting and Data System for thyroid malignancy according to nodule size: a comparison with five society guidelines. Ultrasonography 2021;40:474–85.PubMed
31. Ha EJ, Chung SR, Na DG, Ahn HS, Chung J, Lee JY, et al. 2021 Korean Thyroid Imaging Reporting and data system and imaging-based management of thyroid nodules: Korean Society of Thyroid Radiology consensus statement and recommendations. Korean J Radiol 2021;22:2094–123.Article PubMed PMC PDF
32. Russ G, Bonnema SJ, Erdogan MF, Durante C, Ngu R, Leenhardt L. European Thyroid Association guidelines for ultrasound malignancy risk stratification of thyroid nodules in adults: the EU-TIRADS. Eur Thyroid J 2017;6:225–37.PubMed PMC
33. Tessler FN, Middleton WD, Grant EG, Hoang JK, Berland LL, Teefey SA, et al. ACR Thyroid Imaging, Reporting and Data System (TI-RADS): white paper of the ACR TI-RADS Committee. J Am Coll Radiol 2017;14:587–95.PubMed
34. Zhou J, Yin L, Wei X, Zhang S, Song Y, Luo B, et al. 2020 Chinese guidelines for ultrasound malignancy risk stratification of thyroid nodules: the C-TIRADS. Endocrine 2020;70:256–79.PubMed
35. Gwon HY, Na DG, Noh BJ, Paik W, Yoon SJ, Choi SJ, et al. Thyroid nodules with isolated macrocalcifications: malignancy risk of isolated macrocalcifications and postoperative risk stratification of malignant tumors manifesting as isolated macrocalcifications. Korean J Radiol 2020;21:605–13.Article PubMed PMC PDF
36. Kwon D, Kulich M, Mack WJ, Monedero RM, Joyo E, Angell TE. Malignancy risk of thyroid nodules that are not classifiable by the American Thyroid Association ultrasound risk stratification system: a systematic review and meta-analysis. Thyroid 2023;33:593–602.PubMed PMC
37. Lee EK, Song YS, Kang HC, Kim SW, Na DG, Moon SJ, et al. Korean Thyroid Association guidelines on the management of differentiated thyroid cancers; Part I. Initial management of differentiated thyroid cancers. Chapter 5. Evaluation of recurrence risk postoperatively and initial risk stratification in differentiated thyroid cancer 2024. Int J Thyroidol 2024;17:68–96.Article
38. Jung CK. Reevaluating diagnostic categories and associated malignancy risks in thyroid core needle biopsy. J Pathol Transl Med 2023;57:208–16.PubMed PMC
39. Min HS, Kim JH, Ryoo I, Jung SL, Jung CK. The role of core needle biopsy in the preoperative diagnosis of follicular neoplasm of the thyroid. APMIS 2014;122:993–1000.Article PubMed
40. Kelman AS, Rathan A, Leibowitz J, Burstein DE, Haber RS. Thyroid cytology and the risk of malignancy in thyroid nodules: importance of nuclear atypia in indeterminate specimens. Thyroid 2001;11:271–7.Article PubMed
41. Valderrabano P, Khazai L, Thompson ZJ, Sharpe SC, Tarasova VD, Otto KJ, et al. Cancer risk associated with nuclear atypia in cytologically indeterminate thyroid nodules: a systematic review and meta-analysis. Thyroid 2018;28:210–9.PubMed PMC
42. Lim JX, Nga ME, Chan DK, Tan WB, Parameswaran R, Ngiam KY. Subclassification of Bethesda atypical and follicular neoplasm categories according to nuclear and architectural atypia improves discrimination of thyroid malignancy risk. Thyroid 2018;28:511–21.Article PubMed
43. Usher-Smith JA, Sharp SJ, Griffin SJ. The spectrum effect in tests for risk prediction, screening, and diagnosis. BMJ 2016;353:i3139.PubMed PMC
44. Machens A, Holzhausen HJ, Dralle H. The prognostic value of primary tumor size in papillary and follicular thyroid carcinoma. Cancer 2005;103:2269–73.PubMed
45. Nguyen XV, Roy Choudhury K, Tessler FN, Hoang JK. Effect of tumor size on risk of metastatic disease and survival for thyroid cancer: implications for biopsy guidelines. Thyroid 2018;28:295–300.PubMed

Figure & Data

References

Citations

Citations to this article as recorded by

PubReader
ePub Link

Cite

Cite: export Copy Download Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Risk Stratification of Thyroid Nodules Diagnosed as Follicular Neoplasm on Core Needle Biopsy

DOI: https://doi.org/10.3803/EnM.2024.2256

XML Download

Figure

Risk Stratification of Thyroid Nodules Diagnosed as Follicular Neoplasm on Core Needle Biopsy

Fig. 1. Flow diagram of patient enrollment. US, ultrasound; FNA, fine-needle aspiration; CNB, core needle biopsy; FN, follicular neoplasm.

Fig. 2. Thyroid core needle biopsy diagnostic algorithm for follicular-patterned nodules. Category IIIa is assigned only when the nuclear atypia is evident. If the nodule shows uniformity with equivocal nuclear atypia less than 10%, its category is determined by the criteria of architectural uniformity without nuclear atypia. IIIa, indeterminate follicular lesion with nuclear atypia; IIIb, indeterminate follicular lesion with architectural atypia; IIIc, indeterminate follicular lesion with nuclear and architectural atypia; IIId, indeterminate follicular lesion with oncocytic cell changes; IVa, follicular neoplasm, conventional type; IVb, follicular neoplasm with nuclear atypia; IVc, oncocytic neoplasm.

Fig. 1.

Fig. 2.

Risk Stratification of Thyroid Nodules Diagnosed as Follicular Neoplasm on Core Needle Biopsy

Characteristic	Retrospective dataset	Prospective dataset	P value
No. of patients	195	101
No. of female patients	146 (74.9)	73 (71.9)	0.593
Age, yr	56.9±13.2	56.1±12.8	0.650
No. of nodules	207	106
Maximal nodule diameter, cm^a	2.4 (1.6–3.6)	2.5 (1.7–3.6)	0.945
1–2	78 (37.7)	36 (34.0)	0.788
2.1–3	58 (28.0)	34 (32.1)
3.1–4	29 (14.0)	17 (16.0)
>4	42 (20.3)	19 (17.9)
No. of neoplasm	200 (96.6)	100 (94.3)	0.376
No. of malignant tumor including NIFTP	101 (48.8)	38 (35.8)	0.029
No. of malignant tumor excluding NIFTP	57 (27.5)	22 (20.8)	0.191
Histologic diagnosis by surgery
Benign nodules	106 (51.2)	68 (64.2)	0.029
Thyroid follicular nodular disease	6 (5.7)	6 (8.8)
FA/OA	98 (92.5)	62 (91.2)
Other benign lesions	2 (1.9)	0
Low-risk neoplasm	44 (21.3)	18 (17.0)	0.369
NIFTP	44 (100.0)	16 (88.9)
Thyroid tumors of uncertain malignant potential	0	2 (11.1)
Malignant neoplasm	57(27.5)	20 (18.9)	0.092
PTC	7 (12.3)	9 (45.0)
IE-FVPTC	15 (26.3)	3 (15.0)
FTC/OC	31 (54.4)	7 (35.0)
Other malignant tumors	4 (7.0)	1 (5.0)
High-risk malignant tumors	8 (3.9)	1 (0.9)	0.143
PDTC	4	1
Widely invasive FTC/OC	4^b	0
FTC with vascular invasion (≥4)	1^b	0
Postoperative T4 stage (pT4)	1^b	0
Metastatic cervical lymph node (>3 cm)	0	0
Distant metastasis	0	0

Variable	ROM with NIFTP, %	Univariable analysis		Multivariable analysis		ROM without NIFTP, %	Univariable analysis		Multivariable analysis
Variable	ROM with NIFTP, %	OR (95% CI)	P value	OR (95% CI)	P value	ROM without NIFTP, %	OR (95% CI)	P value	OR (95% CI)	P value
CNB subcategory
IVa	14.8 (4/27)					7.4 (2/27)
IVb	83.7 (87/104)	32.5 (15.1–70.0)	<0.001	39.1 (16.9–90.5)	<0.001	44.2 (46/104)	6.6 (3.2–13.8)	<0.001	6.3 (3.0–13.4)	<0.001
IVc	13.7 (10/73)					12.3 (9/73)
IVd	0.0 (0/3)					0.0 (0/3)
IVa, IVc, IVd	13.6 (14/103)	1.0		1.0		10.7 (11/103)	1.0		1.0
Size, cm
>4.0	57.1 (24/42)	1.3 (0.6–2.8)	0.455	0.9 (0.3–2.8)	0.806	38.1 (16/42)	1.6 (0.7–3.5)	0.268	1.5 (0.6–4.0)	0.370
3.1–4.0	34.5 (10/29)	0.5 (0.2–1.3)	0.155	0.4 (0.1–1.5)	0.164	17.2 (5/29)	0.5 (0.2–1.6)	0.251	0.6 (0.2–2.1)	0.445
2.1–3.0	48.3 (28/58)	0.9 (0.5–1.8)	0.842	1.1 (0.4–3.0)	0.887	24.1 (14/58)	0.8 (0.4–1.8)	0.595	0.9 (0.4–2.3)	0.894
≤2.0	50.0 (39/78)	1.0		1.0		28.2 (22/78)	1.0		1.0
K-TIRADS
5 (high suspicion)	54.8 (17/31)	1.4 (0.6–3.4)	0.394	0.4 (0.1–1.5)	0.172	35.5 (11/31)	1.7 (0.7–4.1)	0.285	1.2 (0.4–3.6)	0.802
4 (intermediate suspicion)	49.1 (53/108)	1.2 (0.6–2.1)	0.652	0.7 (0.3–1.7)	0.417	26.9 (29/108)	1.1 (0.6–2.2)	0.785	0.9 (0.4–2.1)	0.890
2 (benign) or 3 (low suspicion)	45.6 (31/68)	1.0		1.0		25.0 (17/68)	1.0		1.0

Variable	ROM with NIFTP, %	Univariable analysis		Multivariable analysis		ROM without NIFTP, %	Univariable analysis		Multivariable analysis
Variable	ROM with NIFTP, %	OR (95% CI)	P value	OR (95% CI)	P value	ROM without NIFTP, %	OR (95% CI)	P value	OR (95% CI)	P value
CNB subcategory
IVa	29.4 (5/17)					5.9 (1/17)
IVb	55.2 (32/58)	8.6 (3.2–23.4)	<.001	11.7 (3.9–35.2)	<.001	34.5 (20/58)	12.1 (2.7–55.1)	<.001	14.5 (3.0–69.9)	<0.001
IVc	3.3 (1/30)					3.3 (1/30)
IVd	0.0 (0/1)					0.0 (0/1)
IVa, IVc, IVd	12.5 (6/48)	1.0		1.0		4.2 (2/48)	1.0		1.0
Size, cm
>4.0	31.6 (6/19)	0.6 (0.2–2.1)	0.465	0.4 (0.1–2.2)	0.306	26.3 (5/19)	1.3 (0.3–4.5)	0.734	1.0 (0.2–5.8)	0.999
3.1–4.0	47.1 (8/17)	1.2 (0.4–4.0)	0.712	1.3 (0.3–5.8)	0.694	17.6 (3/17)	0.8 (0.2–3.3)	0.702	0.7 (0.1–3.9)	0.673
2.1–3.0	26.5 (9/34)	0.5 (0.2–1.4)	0.184	0.4 (0.1–1.2)	0.099	17.6 (6/34)	0.8 (0.2–2.4)	0.633	0.6 (0.2–2.4)	0.472
≤2.0	41.7 (15/36)	1.0		1.0		22.2 (8/36)	1.0		1.0
K-TIRADS
5 (high suspicion)	36.8 (7/19)	1.0 (0.3–3.3)	0.972	0.3 (0.1–1.5)	0.252	26.3 (5/19)	1.1 (0.3–4.1)	0.868	0.5 (0.1–2.5)	0.373
4 (intermediate suspicion)	35.2 (19/54)	1.0 (0.4–2.3)	0.911	0.5 (0.1–1.7)	0.142	16.7 (9/54)	0.6 (0.2–1.8)	0.39	0.4 (0.1–1.7)	0.211
2 (benign) or 3 (low suspicion)	36.4 (12/33)	1.0		1.0		24.2 (8/33)	1.0		1.0

	Percentage (no./total no) (95% confidence interval)
	Sensitivity	Specificity	PPV	NPV	Accuracy	Surgical rate of benign nodules^a
Retrospective dataset (n=207)
Diagnostic criteria for malignancy with NIFTP
Subcategory IVb	86.1 (87/101)	84.0 (89/106)	83.7 (87/104)	86.4 (89/103)	85.0 (176/207)	16.0 (17/106)
Subcategory IVb	(79.4–92.9)	(77.0–91.0)	(76.6–90.8)	(79.8–93.0)	(80.2–89.9)	(9.1–23.0)
Nodule size >2 cm	61.4 (62/101)	36.8 (39/106)	48.1 (62/129)	50.0 (39/78)	48.8 (101/207)	63.2 (67/106)
Nodule size >2 cm	(51.9–70.9)	(27.6–46.0)	(39.4–56.7)	(38.9–61.1)	(42.0–55.6)	(54.0–72.4)
Nodule size >3 cm	33.7 (34/101)	65.1 (69/106)	47.9 (34/71)	50.7 (69/136)	49.8 (103/207)	34.9 (37/106)
Nodule size >3 cm	(24.5–42.9)	(56.0–74.2)	(36.3–59.5)	(42.3–59.1)	(43.0–56.6)	(25.8–44.0)
Subcategory IVb or size >3 cm	89.1 (90/101)	54.7 (58/106)	65.2 (90/138)	84.1 (58/69)	71.5 (148/207)	45.3 (48/106)
Subcategory IVb or size >3 cm	(83.0–95.2)	(45.2–64.2)	(57.3–73.2)	(75.4–92.7)	(65.4–77.7)	(35.8–54.8)
Diagnostic criteria for malignancy without NIFTP
Subcategory IVb	80.7 (46/57)	61.3 (92/150)	44.2 (46/104)	89.3 (92/103)	66.7 (138/207)	38.7 (58/150)
Subcategory IVb	(70.5–91.0)	(53.5–69.1)	(34.7–53.8)	(83.4–95.3)	(60.2–73.1)	(30.9–46.5)
Nodule size >2 cm	61.4 (35/57)	37.3 (56/150)	27.1 (35/129)	71.8 (56/78)	44.0 (91/207)	62.7 (94/150)
Nodule size >2 cm	(48.8–74.0)	(29.6–45.1)	(19.5–34.9)	(61.8–81.8)	(37.2–50.7)	(54.9–70.4)
Nodule size >3 cm	36.8 (21/57)	66.7 (100/150)	29.6 (21/71)	73.5 (100/136)	58.5 (121/207)	33.3 (50/150)
Nodule size >3 cm	(24.3–49.4)	(59.1–74.2)	(19.0–40.2)	(66.1–80.9)	(51.7–65.2)	(25.8–40.9)
Subcategory IVb or size >3 cm	84.2 (48/57)	40.0 (60/150)	34.8 (48/138)	87.0 (60/69)	52.2 (108/207)	60.0 (90/150)
Subcategory IVb or size >3 cm	(74.7–93.7)	(32.2–47.8)	(26.8–42.7)	(79.0–94.9)	(45.4–59.0)	(52.2–67.8)
Prospective dataset (n=106)
Diagnostic criteria for malignancy with NIFTP
Subcategory IVb	84.2 (32/38)	61.8 (42/68)	55.2 (32/58)	87.5 (42/48)	69.8 (74/106)	38.2 (26/68)
Subcategory IVb	(72.6,95.8)	(50.2–73.3)	(42.4–68.0)	(78.1–96.9)	(61.1–78.6)	(26.7–49.8)
Nodule size >2 cm	60.5 (23/38)	30.9 (21/68)	32.9 (23/70)	58.3 (21/36)	41.5 (44/106)	69.1 (47/68)
Nodule size >2 cm	(45.0–76.1)	(19.9–41.9)	(21.9–43.9)	(42.2–74.4)	(32.1–50.9)	(58.1–80.1)
Nodule size >3 cm	36.8 (14/38)	67.6 (46/68)	38.9 (14/36)	65.7 (46/70)	56.6 (60/106)	32.4 (22/68)
Nodule size >3 cm	(21.5–52.2)	(56.5–78.8)	(23.0–54.8)	(54.6–76.8)	(47.2–66.0)	(21.2–43.5)
Subcategory IVb or size >3 cm	94.7 (36/38)	39.7 (27/68)	46.8 (36/77)	93.1 (27/29)	59.4 (63/106)	60.3 (41/68)
Subcategory IVb or size >3 cm	(87.6–100.0)	(28.1–51.3)	(35.6–57.9)	(83.9–100.0)	(50.1–68.8)	(48.7–71.9)
Diagnostic criteria for malignancy without NIFTP
Subcategory IVb	90.9 (20/22)	54.8 (46/84)	34.5 (20/58)	95.8 (46/48)	62.3 (66/106)	45.2 (38/84)
Subcategory IVb	(78.9–100.0)	(44.1–65.4)	(22.3–46.7)	(90.2–100.0)	(53.0–71.5)	(34.6–55.9)
Nodule size >2 cm	63.6 (14/22)	33.3 (28/84)	20.0 (14/70)	77.8 (28/36)	39.6 (42/106)	66.7 (56/84)
Nodule size >2 cm	(43.5–83.7)	(23.3–43.4)	(10.6–29.4)	(64.2–91.4)	(30.3–48.9)	(56.6–76.8)
Nodule size >3 cm	36.4 (8/22)	66.7 (56/84)	22.2 (8/36)	80.0 (56/70)	60.4 (64/106)	33.3 (28/84)
Nodule size >3 cm	(16.3–56.5)	(56.6–76.8)	(8.6–35.8)	(70.6–89.4)	(51.1–69.7)	(23.3–43.4)
Subcategory IVb or size >3 cm	100.0 (22/22)	34.5 (29/84)	28.6 (22/77)	100.0 (29/29)	48.1 (51/106)	65.5 (55/84)
Subcategory IVb or size >3 cm	(100.0–100.0)	(24.4–44.7)	(18.5–38.7)	(100.0–100.0)	(38.6–57.6)	(55.3,75.6)

Table 1. Demographic Patient Data and the Characteristics of Thyroid Nodules in Retrospective and Prospective Datasets

Values are expressed as number (%), mean±standard deviation, or median (interquartile range).

Measured on ultrasound;

Same one patient overlaps.

Table 2. Risk of Malignancy and Association of Core Needle Biopsy Subcategories, Nodule Size, K-TIRADS Categories with Malignancy in Retrospective Dataset (n=207)

Data in parentheses are raw data.

Table 3. Risk of Malignancy and Association of Core Needle Biopsy Subcategories, Nodule Size, K-TIRADS Categories with Malignancy in Prospective Dataset (n=106)

Data in parentheses are raw data.

Table 4. Diagnostic Performance of Core Needle Biopsy Subcategory (IVb), Nodule Size, and Combined Criteria for the Diagnosis of Malignancy

Data in parentheses are raw data.

PPV, positive predictive value; NPV, negative predictive value; NIFTP, noninvasive follicular thyroid neoplasm with papillary-like nuclear feature.

Articles

ABSTRACT

INTRODUCTION

METHODS

RESULTS

DISCUSSION

Supplementary Material

Supplemental Table S1.

Supplemental Table S2.

Supplemental Table S3.

Article information

References

Figure & Data

References

Citations

Fig. 1.

Fig. 2.

Table 1.

Table 2.

Table 3.

Table 4.