Erratum: Introduction of a New Staging System of Breast Cancer for Radiologists: An Emphasis on the Prognostic Stage.

This corrects the article on p. 69 in vol. 20, PMID: 30627023.

Deep Learning for the Detection of Breast Cancers on Chest Computed Tomography.

BACKGROUND: Incidental breast cancers can be detected on chest computed tomography (CT) scans. With the use of deep learning, the sensitivity of incidental breast cancer detection on chest CT would improve. This study aimed to evaluate the performance of a deep learning algorithm to detect breast cancers on chest CT and to validate the results in the internal and external datasets. PATIENTS AND METHODS: This retrospective study collected 1170 preoperative chest CT scans after the diagnosis of breast cancer for algorithm development (n = 1070), internal test (n = 100), and external test (n = 100). A deep learning algorithm based on RetinaNet was developed and tested to detect breast cancer on chest CT. RESULTS: In the internal test set, the algorithm detected 96.5% of breast cancers with 13.5 false positives per case (FPs/case). In the external test set, the algorithm detected 96.1% of breast cancers with 15.6 FPs/case. When the candidate probability of 0.3 was used as the cutoff value, the sensitivities were 92.0% with 7.36 FPs/case for the internal test set and 93.0% with 8.85 FPs/case for the external test set. When the candidate probability of 0.4 was used as the cutoff value, the sensitivities were 88.5% with 5.24 FPs/case in the internal test set and 90.7% with 6.3 FPs/case in the external test set. CONCLUSION: The deep learning algorithm could sensitively detect breast cancer on chest CT in both the internal and external test sets.

Comparison of Abbreviated MRI with Mammography and Ultrasound in Women with a Personal History of Breast Cancer.

RATIONALE AND OBJECTIVES: To compare abbreviated MRI with mammography and US for screening in women with a personal history of breast cancer. In addition, the first and subsequent rounds of abbreviated MRI were compared. MATERIALS AND METHODS: The Institutional Review Board approved this retrospective study. Nine hundred and thirty-nine abbreviated MRI scans of 710 women with a personal history of breast cancer were included (mean age, 54.1±9.4 years). The diagnostic performances of abbreviated MRI, mammography, and US for the detection of the second breast cancer were compared. When more than one round of abbreviated MRI was performed, we compared the scans of the first and subsequent rounds. RESULTS: There were 15 (2.1%) cases of second breast cancer. Thirty-nine of the 939 abbreviated MRI scans were diagnosed as positive; of them, 11 were diagnosed as breast cancer, with a PPV2 of 28.2% (US, 19.0%; mammography, 28.6%). The AUC, sensitivity, and cancer detection rate (per 1000) were the highest for abbreviated MRI (0.829, 68.8% and 11.7) (US: 0.616, 25.0%, and 4.3; mammography: 0.560, 12.5%, and 2.1, respectively). Two hundred women underwent multiple rounds of abbreviated MRI. AUC, PPV2, and the cancer detection rates were higher during the subsequent rounds (0.987, 45.5, and 21.8) than during the first round (0.605, 11.1, and 5). CONCLUSION: Abbreviated MRI for women with a personal history of breast cancer was more sensitive for the diagnosis of second breast cancer than US or mammography. In addition, subsequent rounds of MRI showed better performance than the first round.

The usefulness of ultrafast MRI evaluation for predicting histologic upgrade of ductal carcinoma in situ.

PURPOSE: The purpose of this study was to investigate the usefulness of ultrafast MRI with conventional dynamic contrast-enhanced (DCE)-MRI for predicting histologic upgrade of ductal carcinoma in situ (DCIS) to invasive cancer. METHODS: This retrospective study enrolled 53 biopsy-proven DCIS lesions in 53 patients and divided into two groups based on postoperative histopathologic diagnoses: non-upgrade and upgrade to invasive cancer groups. Imaging features of conventional DCE-MRI and ultrafast MRI, and histopathologic features were reviewed and compared between the two groups. Interobserver agreements for MRI features were analyzed by two radiologists. The radiologic and histopathologic parameters for predicting histologic upgrade of DCIS were identified using multiple linear regression. RESULTS: Seventeen lesions (32.1 %) were histologically upgraded to invasive cancer after surgery. The interobserver agreement for ultrafast MRI parameters was excellent, and maximum slope (MS) and maximum enhancement (ME) showed the highest reliability (intraclass correlation coefficients, 0.907 and 0.897, respectively). The upgrade group showed significantly larger lesion size on MRI (median 40 mm [25th to 75th percentiles 16.0-83.0] vs. 18.5 mm [10.0-29.8], p < 0.001), higher MS (12.1 %/s [8.2-13.9] vs. 8.7 %/s [6.4-11.1], p = 0.004), and higher ME (236.5 % [153.7-253.7] vs. 175.4 % [140.1-207.7], p = 0.027) than non-upgrade group. Lesion size (≥ 20 mm), MS (> 11.5 %), and ME (> 229.1 %) were significant predictors for histologic upgrade, which could predict 10 cases of histologic upgrade (10/17, 58.8 %) without a false-positive case. CONCLUSION: Preoperative ultrafast MRI with conventional DCE-MRI could be useful in management decisions for DCIS patients.

Ultrasonography-Based Radiomics of Screening-Detected Ductal Carcinoma In Situ According to Visibility on Mammography.

ABSTRACT: Ductal carcinoma in situ (DCIS) has different prognostic factors according to the detection modality. The purpose of this study was to compare parameters from a radiomic analysis of ultrasonography (US) images for DCIS detected on screening mammography (MMG) and US and detected on screening US only. A total of 154 surgically confirmed DCIS visible on US were included. Regions of interest were drawn onto US images of DCIS, and texture analysis was performed. Lesions were classified into those detected by both US and MMG (the US-MMG group) and those detected by US only (the US group). Analysis parameters were compared between the US-MMG group and the US group. Ninety-six lesions were included in the US-MMG group and 58 lesions in the US group. Energy, entropy, maximum, mean absolute deviation, range, SD, and variance were significantly higher in the US-MMG group than the US group. Kurtosis, skewness, and uniformity were significantly lower in the US-MMG group than the US group. Among the 22 gray-level cooccurrence matrix parameters, 18, 21, 22, 20, and 21 parameters were significantly different between the 2 groups in 0, 45, 90, and 135 degrees and the average value. Among the 11 gray-level run-length matrix parameters, 6, 6, 7, 7, and 6 parameters were significantly different in 0, 45, 90, and 135 degrees and the average value. Inverse variance and gray-level nonuniformity were the most different features between the 2 groups. Screening-detected DCIS showed different radiomic features according to the detection modality.

Diagnosis of thyroid nodules on ultrasonography by a deep convolutional neural network.

The purpose of this study was to evaluate and compare the diagnostic performances of the deep convolutional neural network (CNN) and expert radiologists for differentiating thyroid nodules on ultrasonography (US), and to validate the results in multicenter data sets. This multicenter retrospective study collected 15,375 US images of thyroid nodules for algorithm development (n = 13,560, Severance Hospital, SH training set), the internal test (n = 634, SH test set), and the external test (n = 781, Samsung Medical Center, SMC set; n = 200, CHA Bundang Medical Center, CBMC set; n = 200, Kyung Hee University Hospital, KUH set). Two individual CNNs and two classification ensembles (CNNE1 and CNNE2) were tested to differentiate malignant and benign thyroid nodules. CNNs demonstrated high area under the curves (AUCs) to diagnose malignant thyroid nodules (0.898-0.937 for the internal test set and 0.821-0.885 for the external test sets). AUC was significantly higher for CNNE2 than radiologists in the SH test set (0.932 vs. 0.840, P < 0.001). AUC was not significantly different between CNNE2 and radiologists in the external test sets (P = 0.113, 0.126, and 0.690). CNN showed diagnostic performances comparable to expert radiologists for differentiating thyroid nodules on US in both the internal and external test sets.

Three-dimensional radiomics of triple-negative breast cancer: Prediction of systemic recurrence.

This paper evaluated 3-dimensional radiomics features of breast magnetic resonance imaging (MRI) as prognostic factors for predicting systemic recurrence in triple-negative breast cancer (TNBC) and validated the results with a different MRI scanner. The Rad score was generated from 3-dimensional radiomic features of MRI for 231 TNBCs (training set (GE scanner), n = 182; validation set (Philips scanner), n = 49). The Clinical and Rad models to predict systemic recurrence were built up and the models were externally validated. In the training set, the Rad score was significantly higher in the group with systemic recurrence (median, -8.430) than the group without (median, -9.873, P < 0.001). The C-index of the Rad model to predict systemic recurrence in the training set was 0.97, which was significantly higher than in the Clinical model (0.879; P = 0.009). When the models were externally validated, the C-index of the Rad model was 0.848, lower than the 0.939 of the Clinical model, although the difference was not statistically significant (P = 0.100). The Rad model for predicting systemic recurrence in TNBC showed a significantly higher C-index than the Clinical model. However, external validation with a different MRI scanner did not show the Rad model to be superior over the Clinical model.

MRI diagnostic features for predicting nipple-areolar-complex involvement in breast cancer.

PURPOSE: To evaluate magnetic resonance imaging (MRI) features including tumor-nipple enhancement for predicting nipple-areolar-complex (NAC) involvement in breast cancer. METHODS: A total of 249 breast cancer cases with available preoperative MRI data were studied. On MRI, tumor-nipple enhancement was evaluated 2 min (early) and 6 min (delayed) after contrast injection. Asymmetric nipple enhancement, mass size, tumor-nipple distance, presence of nipple discharge, and enhancement type (mass or non-mass enhancement) were evaluated. RESULTS: Of the 249 breast cancers, 24 (9.6%) showed NAC involvement. Early tumor-nipple enhancement was found in 37 (14.9%) cancers and delayed tumor-nipple enhancement in 12 (4.8%) cancers. On comparing early and delayed enhancement, 43.2% (16/37) of early tumor-nipple enhancements showed NAC involvement while 25.0% (3/12) of delayed tumor-nipple enhancements showed NAC involvement (P = 0.323). On multivariate analysis, tumor-nipple enhancement (P = 0.004) and tumor-nipple distance with 2 cm cut-off (P = 0.044) were significant. The area under the receiver-operating-characteristic curve (AUC) of early tumor-nipple enhancement was significantly higher (0.787) than that of delayed tumor-nipple enhancement (0.542, P = 0.001). The AUC was the highest when any kind of tumor-nipple enhancement was combined with a tumor-nipple distance of 1 cm (0.878). CONCLUSIONS: Tumor-nipple enhancement and tumor-nipple distance on MRI could predict NAC involvement in breast cancer. When enhancement was evaluated on both early and delayed phase images with a combined tumor-nipple distance of ≤1 cm, the prediction of NAC involvement showed the best performance.

Outcomes of Ductal Carcinoma In Situ According to Detection Modality: A Multicenter Study Comparing Recurrence Between Mammography and Breast US.

The purpose of this study was to determine whether disease recurrence and intrinsic characteristics of ductal carcinoma in situ (DCIS) are associated with the imaging method of detection in asymptomatic women. This multicenter, retrospective study included 844 women treated for asymptomatic DCIS who had pre-operative mammography and breast ultrasonography (US) studies available. Of the 844 women, 25 (3.0%) developed recurrences. Patients in the US group had significantly lower 5- and 10-y recurrence-free survival (RFS) rates compared with patients in the mammography group (p = 0.011). US-detected DCIS showed significantly lower 5-and 10-y RFS rates compared with mammography-detected DCIS in patients <50 y or with mammographically dense breasts (p = 0.002 and 0.002, respectively). US as the detection modality (hazard ratio [HR]: 4.451; 95% confidence interval [CI]: 1.530, 12.950; p = 0.006) and HER2 positivity (HR: 4.036; 95% CI: 1.438; 11.330; p = 0.008) were significantly associated with recurrence. We concluded that US as the detection modality and HER2 positivity were significantly associated with recurrence in patients treated for asymptomatic DCIS.

Can enhancement types on preoperative MRI reflect prognostic factors and surgical outcomes in invasive breast cancer?

OBJECTIVES: This study was conducted in order to evaluate whether enhancement types on preoperative MRI can reflect prognostic factors and surgical outcomes in invasive breast cancer. METHODS: Among 484 consecutive patients who underwent preoperative breast MRI from October 2014 to July 2017 for biopsy-proven breast cancer, 313 patients with 315 invasive breast cancers who underwent subsequent surgery were finally included in this study. Two radiologists retrospectively reviewed preoperative MRI findings of these 315 lesions and categorized them to mass, nonmass, and combined type according to enhancement features. Combined type was defined as coexisted mass and nonmass enhancement. Histopathologic results focusing on prognostic factors and surgical outcomes were compared among the three types of lesion using Pearson's chi-square, linear-by-linear association, Kruskal-Wallis, one-way ANOVA test, and multinomial logistic regression. RESULTS: Of the cancers analyzed, 198 (62.9%) were mass, 59 (18.7%) were nonmass, and 58 (18.4%) were combined type. The nonmass type showed the smallest invasive tumor size (p < 0.001) and the most common positive HER2 receptor status (p = 0.001). The combined type had the most frequent lymphovascular invasion (p = 0.011), axillary lymph node-positive status (p = 0.031), operation changes (p < 0.001), and first resection margin-positive status (p < 0.001). Initial operation of mastectomy was more frequent in the nonmass and combined types than that in the mass type (p < 0.001). But HER2 receptor status and operation changes showed no statistical significance on multivariate analysis. CONCLUSIONS: Enhancement types on preoperative MRI reflect different prognostic factors and surgical outcomes in invasive breast cancer. KEY POINTS: • Morphologic features of contrast media uptake on contrast-enhanced MRI may be related with fundamental biological differences of invasive breast cancers. • Mass or nonmass enhancement type on preoperative MRI might reflect different prognostic factors and surgical outcomes in invasive breast cancer. • The combined mass and nonmass enhancement type might be associated with poorer prognosis and worse surgical outcomes.

Introduction of a New Staging System of Breast Cancer for Radiologists: An Emphasis on the Prognostic Stage.

In 2017, the American Joint Committee on Cancer announced the 8th edition of its cancer staging system. For breast cancer, the most significant change in the staging system is the incorporation of biomarkers into the anatomic staging to create prognostic stages. Different prognostic stages are assigned to tumors with the same anatomic stages according to the tumor grade, hormone receptor (estrogen receptor; progesterone receptor) status, and HER2 status. A Clinical Prognostic Stage is assigned to all patients regardless of the type of therapy used; in contrast, a Pathologic Prognosis Stage is assigned to patients in whom surgery is the initial treatment. In a few situations, low Oncotype DX recurrence scores can change the prognostic stage. The radiologists need to understand the importance of the biologic factors that can influence cancer staging.

Role of elastography for downgrading BI-RADS category 4a breast lesions according to risk factors.

BACKGROUND: Elastography has been introduced as an additional diagnostic tool to ultrasonography (US) which helps clinicians decide whether or not to perform biopsy on US-detected lesions. PURPOSE: To evaluate the role of strain elastography in downgrading Breast Imaging Reporting and Data System (BI-RADS) category 4a breast lesions according to personal risk factors for breast cancer in asymptomatic women. MATERIAL AND METHODS: Strain elastography features of a total of 255 asymptomatic category 4a lesions were classified as soft and not soft (intermediate and hard). Malignancy was confirmed by surgery or biopsy, and benignity was confirmed by surgery or biopsy with no change on US for at least six months. Malignancy rates of lesions with soft and not soft elastography were calculated according to the presence of risk factors. RESULTS: Of 255 lesions, 25 (9.8%) were malignant and 230 (90.2%) were benign. Of 195 lesions in average-risk women, the malignancy rate of lesions with soft elastography was 1.5% (1/68), which was significantly lower than the 14.2% (18/127) of lesions with not soft elastography ( P = 0.004). Of 60 lesions in increased-risk women, the malignancy rate of lesions with soft elastography was 15.0% (3/20), which was not significantly different from the 7.5% (3/40) of lesions with not soft elastography ( P = 0.390). CONCLUSION: In average-risk women, category 4a lesions with soft elastography could be followed up with US because of a low malignancy rate of 1.5%.

New Doppler imaging technique for assessing angiogenesis in breast tumors: correlation with immunohistochemically analyzed microvessels density.

BACKGROUND: Microvessel density (MVD) is associated with grade and prognosis in breast tumors. However, conventional color Doppler flow (CDF) imaging has been limited to represent MVD of breast tumors. PURPOSE: To evaluate whether a new Doppler imaging technique (AngioPLUS) can represent MVD of breast tumors. MATERIAL AND METHODS: The institutional review board approved this retrospective study, and patients' informed consent was waived. CDF and AngioPLUS were available in pathologically confirmed 55 breast tumors of 53 women. For each lesion, vascular flow patterns (distribution and amount) of both Doppler images were retrospectively reviewed, and MVD was measured using immunohistochemical analysis of the biopsied tissue sections. MVD was subcategorized as low or high group with reference to the median. The associations between the Doppler features and MVD were evaluated using Fisher's exact test and Student's t test. RESULTS: Of the 55 masses, 28 (50.9%) were benign and 27 (49.1%) were malignant. Vascular flow distribution and amount of both Doppler imaging were different between the benign and malignant lesions (CDF, P = 0.020 and P = 0.010; AngioPLUS, P = 0.002 and P = 0.005). MVD had no significant relationships with CDF features, but vascular flow distribution on AngioPLUS showed significant differences between the lesions with low and high MVD ( P = 0.020); Combined distribution was more frequent in the high MVD lesions than in the low MVD lesions (17/28, 60.7% vs. 6/27, 22.2%). CONCLUSION: Our data confirmed the correlation between a new Doppler imaging technique, AngioPLUS, and MVD. We suggest that AngioPLUS can be used for assessing MVD in breast tumors.

Comparison of the Diagnostic Performance of Power Doppler Ultrasound and a New Microvascular Doppler Ultrasound Technique (AngioPLUS) for Differentiating Benign and Malignant Breast Masses.

This study was performed to compare the diagnostic performance of power Doppler ultrasound (US) and a new microvascular Doppler US technique (AngioPLUS; SuperSonic Imagine, Aix-en-Provence, France) for differentiating benign and malignant breast masses. Power Doppler US and AngioPLUS findings were available in 124 breast masses with confirmed pathologic results (benign, 80 [64.5%]; malignant, 44 [35.5%]). The diagnostic performance of each tool was calculated to distinguish benign from malignant masses using a receiver operating characteristic curve analysis and compared. The area under the curve showed that AngioPLUS was superior to power Doppler US in differentiating benign from malignant breast masses, but the difference was not statistically significant.

Diagnostic performances and interobserver agreement according to observer experience: a comparison study using three guidelines for management of thyroid nodules.

Background The differences regarding categorization of thyroid nodules among different guidelines may affect the diagnostic performances and agreement among observers. Purpose To compare the diagnostic performances and agreements between observers with various degree of experience when applying different guidelines for stratifying thyroid nodules using suspicious ultrasonography (US) features. Material and Methods This retrospective study included 370 thyroid nodules (≥10 mm). Four observers, grouped as experienced and inexperienced, evaluated the US features and made final assessments according to the Kim criteria, Thyroid Imaging Reporting and Data System (TIRADS) by Kwak et al., and the 2015 American Thyroid Association (ATA) guideline. Diagnostic performances and agreements among the two groups were compared. Results The Kim criteria shows higher specificity with significantly lower sensitivity when compared to TIRADS and the 2015 ATA guideline (all P < 0.001), regardless of the level of experience. The experienced group showed significantly higher specificity with the Kim criteria and the 2015 ATA guideline compared to the inexperienced group ( P < 0.001), and the inexperienced group showed significantly higher sensitivity using the Kim criteria ( P = 0.002). The experienced group showed significantly higher agreement than the inexperienced group when using TIRADS while higher agreement was seen when using the 2015 ATA guideline for the inexperienced group. Agreement was not significantly different for the Kim criteria according to observer experience. Conclusion The diagnostic performances and agreements show significant differences in risk stratification of thyroid nodules according to the three guidelines using suspicious US features and the level of experience of the observer.

Intrinsic Subtypes of Breast Cancers Initially Assessed as Probably Benign or of Low Suspicion on Ultrasonography Differ According to Tumor Size.

OBJECTIVES: To investigate whether the intrinsic subtypes of breast cancers initially assessed as American College of Radiology Breast Imaging and Reporting System (BI-RADS) category 3 or 4a differ according to tumor size. METHODS: A total of 444 breast cancers in 439 patients initially assessed as BI-RADS 3 or 4a through ultrasound-guided core needle biopsy were included. Tumors were classified by the size criterion of 10 mm or smaller or larger than 10 mm and categorized as the luminal type (estrogen receptor [ER] positive and/or progesterone receptor [PR] positive and human epidermal growth factor receptor 2 [HER2] negative), HER2 type (HER2 positive regardless of ER or PR status), or triple-negative type (ER negative, PR negative, and HER2 negative). The relationships between tumor size and breast cancer intrinsic subtypes were analyzed. RESULTS: A total of 247 (55.6%) cancers were 10 mm or smaller, and 197 (44.4%) were larger than 10 mm. The luminal type was more frequently present in tumors of 10 mm or smaller (185 of 247 [74.9%]) than those larger than 10 mm (112 of 197 [56.9%]; P < .001). No significant difference was noted in the frequency of the HER2 type in tumors of 10 and mm smaller and those larger than 10 mm (27 of 247 [10.9%] versus 36 of 197 [18.3%]; P = .084). The triple-negative type was more frequently present in tumors larger than 10 mm than those of 10 mm or smaller (49 of /197 [24.9%] versus 35 of 247 [14.2%]; P = .012). CONCLUSIONS: Breast cancers assessed as BI-RADS category 3 or 4a had differing intrinsic subtypes according to tumor size, as the luminal type was more frequently present in tumors of 10 mm or smaller than those larger than 10 mm, whereas the triple-negative type was more frequently present in tumors larger than 10 mm.

Additional Magnetic Resonance Imaging-Detected Suspicious Lesions in Known Patients With Breast Cancer: Comparison of Second-Look Digital Tomosynthesis and Ultrasonography.

PURPOSE: The aim of the study was to compare the performances of second-look digital breast tomosynthesis (DBT) and ultrasonography (US) for additional magnetic resonance imaging (MRI)-detected suspicious lesions in patients referred for breast cancer. METHODS: Fifty-five patients (52.5 [9.3] years) with breast cancers newly diagnosed at a referring hospital underwent MRI and afterward DBT and US in our institution. The frequency for detecting additional MRI-detected suspicious lesions on DBT and US and performances of DBT and US were compared. RESULTS: Of 37 additional MRI-detected suspicious lesions, 27 were detected on DBT and/or US; 2 were detected on both DBT and US, 1 was detected only on DBT, and 24 were detected only on US. Ultrasonography detected more additional suspicious lesions than DBT (P < 0.0001). Twelve (44.4%) of the 27 lesions were malignant. The sensitivity and negative predictive value of US (100.0% and 100.0%) were significantly higher than those of DBT (16.7% and 83.6%, P < 0.001 and P = 0.001). The specificity of DBT was significantly higher (98.1%) than that of US (78.9%, P = 0.002). Positive predictive values of DBT and US were not significantly different (66.7% vs 52.2%, P = 0.598). The area under the receiver operating characteristic curve of US was significantly higher (0.894) than that of DBT (0.574, P < 0.001). CONCLUSIONS: Second-look US showed higher performances than DBT for detecting and characterizing suspicious lesions additionally detected on MRI in patients referred for breast cancer.

Comparison of Ultrasound, Pathologic and Prognostic Characteristics of the Follicular Variant of Papillary Thyroid Cancer According to Fine-Needle Aspiration Cytology.

The aim of the study described here was to compare ultrasound features, pathologic characteristics and prognoses of the follicular variant of papillary thyroid carcinoma (FVPTC) according to cytology results. Eighty-seven FVPTCs were classified according to the first cytology results as the surgery group (n = 66, follicular neoplasm/Hürthle cell neoplasm, suspicious for malignancy and malignancy) and the indeterminate group (n = 21, non-diagnostic and benign), for whom the management was follow-up. The indeterminate group had a longer mean interval to surgery (p = 0.020) and larger tumor size (p = 0.018). More tumors were assessed as probably benign in the indeterminate group than in the surgery group (p < 0.001). Extrathyroidal extension and lymph node metastasis did not significantly differ between the two groups, and no patient had a recurrence (mean follow-up interval: 54.9 ± 16.9 mo). The indeterminate group exhibited more probably benign features and larger size on ultrasound, with surgery being performed at a later date. However, aggressive pathologic characteristics and tumor recurrence did not significantly differ between the two groups.

Repeat fine-needle aspiration can be performed at 6 months or more after initial atypia of undetermined significance or follicular lesion of undetermined significance results for thyroid nodules 10 mm or larger.

OBJECTIVES: To investigate whether repeat ultrasound-guided fine-needle aspiration (US-FNA) in initial atypia of undetermined significance or follicular lesion of undetermined significance (AUS/FLUS) results could be performed 6 months after or more. METHODS: A total of 221 AUS/FLUS ≥10 mm with any follow-up were grouped according to the first follow-up interval at less than 6 months (group 1, n = 87) and 6 months or more (group 2, n = 134). Clinical features, final assessment of ultrasound (US) or the Thyroid Imaging Reporting and Data System (TIRADS), tumour size, extrathyroidal extension and lymph node metastasis in malignancies were compared. RESULTS: Thirty-four (15.4 %) were malignant. Age, gender, size, final assessment, TIRADS and malignancy rate were not significantly different between the two groups (p = 0.660, 0.691, 0.502, 0.237, 0.819 and 0.420). Tumour size, extrathyroidal extension and lymph node metastasis were not significantly different between the two malignancy groups (p = 0.770, 0.611 and 0.068). Two of 10 nodules with increased size were malignancies found at 7.1 and 25.0 months. None of 33 nodules (14.9 %) with decreased size at a median 10 months were malignant. CONCLUSIONS: Repeat US-FNA performed on nodules ≥10 mm at 6 months or more after initial AUS/FLUS results can reduce unnecessary repeat US-FNAs without progression of malignancy. KEY POINTS: • Follow-up intervals of AUS/FLUS did not affect the malignancy rate • Tumour stage was not different according to the follow-up intervals • None of the nodules with decreased size were malignant • Repeat US-FNA can be performed at ≥6 months after initial AUS/FLUS.