BRAF Mutation Analysis and Sonography as Adjuncts to Fine- Needle Aspiration Cytology of Papillary Thyroid Carcinoma: Their Relationships and Roles

Transcription

1 Neuroradiology/Head and Neck Imaging Original Research Moon et al. BRAF Analysis and Sonography to Diagnose PTC Neuroradiology/Head and Neck Imaging Original Research Won-Jin Moon 1 Nami Choi 1 Jin Woo Choi 1 Suk Kyeong Kim 2 Tae Sook Hwang 3 Moon WJ, Choi N, Choi JW, Kim SK, Hwang TS Keywords: BRAF mutation, fine-needle biopsy, papillary carcinoma, protooncogene proteins, sonography, thyroid neoplasms DOI: /AJR Received May 10, 2011; accepted without revision July 24, This work was supported by Konkuk University. 1 Department of Radiology, Konkuk University Medical Center, Konkuk University School of Medicine, 4-12, Hwayang-dong, Gwangjin-gu, Seoul , Korea. Address correspondence to W. J. Moon (mdmoonwj@naver.com). 2 Department of Internal Medicine, Konkuk University Medical Center, Konkuk University School of Medicine, Seoul, Korea. 3 Department of Pathology, Konkuk University Medical Center, Konkuk University School of Medicine, Seoul, Korea. AJR 2012; 198: X/12/ American Roentgen Ray Society BRAF Mutation Analysis and Sonography as Adjuncts to Fine- Needle Aspiration Cytology of Papillary Thyroid Carcinoma: Their Relationships and Roles OBJECTIVE. The objective of our study was to evaluate the relationships between BRAF mutation status, sonography findings, and fine-needle aspiration cytology features in patients with papillary thyroid carcinoma (PTC) and to evaluate the diagnostic merits of BRAF mutation status and sonography findings as adjuncts to cytologic diagnoses. MATERIALS AND METHODS. From March 2006 through June 2008, clinicopathologic factors, sonography findings, cytology results, and BRAF mutation status were evaluated in 524 patients (437 women and 87 men) with 553 thyroid nodules; of the 170 malignant nodules, 164 were PTCs. Clinicopathologic factors, sonography findings, and cytology results were correlated with BRAF status. The diagnostic sensitivities and specificities of sonography, cytology, and BRAF analysis and their combinations were compared. RESULTS. The V600E mutation of BRAF (BRAF V600E ) was detected in 141 of 170 malignant thyroid nodules (82.9%) (140 PTCs and one follicular variant of PTC). Multiple logistic regression revealed that BRAF status was not associated with sonography features with the exception of a negative relation between BRAF V600E and an irregular shape (p = 0.004). An indeterminate cytology result was more frequent for BRAF-negative PTC than BRAF-positive PTC (p = 0.035). By adding BRAF status to cytology, diagnostic sensitivity for PTC was significantly increased (94.1%) as compared with cytology alone (81.8%) (p < 0.001). The triple combination that is, sonography, cytology, and BRAF analysis showed higher sensitivity than BRAF plus cytology (98.2% vs 94.1%, respectively) (p < 0.05). CONCLUSION. The sonography features of PTC, other than an irregular shape, are not related to BRAF status and the combination of sonography and BRAF testing would increase the diagnostic accuracy of cytologic diagnoses of PTC. S ince the advent of high-resolution sonography and its widened use, the diagnosis of thyroid carcinomas has increased rapidly worldwide. The ubiquitous presence of benign thyroid nodules has further increased the use of sonography for preoperative differentiation and for the guidance of fine-needle aspiration (FNA) for cytologic diagnosis. Studies have consistently shown that sonography can distinguish between benign and malignant nodules with high sensitivity but with relatively low specificity [1 3]. FNA is regarded as a reference standard for diagnosing thyroid malignancies preoperatively because of its high specificity, but FNA has a relatively high rate of indeterminate cytology findings, which vary from 15% to 25% [4, 5]. A number of genetic analyses have been found to enhance the diagnostic accuracy of cytology and as a predictor of poor progno- sis in patients with thyroid carcinomas. The most promising of these is analysis for the V600E mutation of BRAF (BRAF V600E ) [6, 7]. The prevalence of BRAF V600E has been reported to be 29 69% in the West [6] and as high as 73 86% in Korea [8 10]. BRAF V600E has been reported to be present exclusively in malignant thyroid nodules of papillary thyroid carcinoma (PTC) [6, 7]. Few recent studies have been conducted about the relationship between BRAF V600E status and sonography features in PTC, and those conducted have produced inconsistent results [11, 12]. The results of several studies have suggested that BRAF V600E status would be helpful for diagnosing the status of nodules with an indeterminate cytology results [7, 11, 13, 14]. Interestingly, in a recent study, investigators found that sonography findings could be helpful for diagnosing malignancy in 668 AJR:198, March 2012

2 BRAF Analysis and Sonography to Diagnose PTC BRAF V600E -negative nodules [15]. Furthermore, it has been suggested that sonography features may be helpful for improving the diagnosis of nodules with an indeterminate cytology result by FNA [16]. Hence, in this study, we evaluated retrospectively the relationship between BRAF mutation status, sonography findings, and FNA cytology features in Korean patients with PTC. In addition, we evaluated the roles of BRAF mutation status and sonography findings as adjuncts to FNA cytologic diagnosis. Materials and Methods This retrospective study was approved by our institutional review board. We obtained written informed consent from all patients before performing the ultrasound-guided FNA procedure and the genetic analysis. Patient Selection From March 2006 through June 2008, 569 consecutive patients with 598 nodules underwent sonography, sonography-guided FNA, and BRAF analysis. Of the 569 patients, only patients who met the following criteria were included: those who underwent surgery after thyroid sonography and FNA, those who underwent FNA at least twice with a 1-year interval for a benign thyroid lesion, and those who underwent FNA and sonography follow-up (> 12 months after FNA cytology diagnosis) for a benign thyroid lesion. As a result, 37 nodules were excluded because of an indeterminate cytology diagnosis without followup and eight nodules were excluded because of insufficient cytology without follow-up. Accordingly, 524 patients (437 women, 87 men; mean age, 50 years; range, years) with 553 thyroid nodules were included for analysis. All 170 malignant nodules were confirmed at surgery, whereas the 383 benign nodules were confirmed at surgery (n = 8), by repeated sonography-guided FNA (n = 27), or by sonography-guided FNA and sonography during follow-up (n = 348). The final diagnoses of malignant thyroid nodules (n = 170) were as follows: conventional PTC (n = 161, 94.7%), a follicular variant of PTC (n = 3, 1.8%), and follicular carcinoma (n = 6, 3.5%). The nodules ranged in size from 2.5 to 80 mm (mean ± SD, 14.2 ± 11.2 mm). Multifocality, extracapsular invasion, lymph node metastasis (central and lateral), and tumor stages were recorded for patients with PTC. Sonographic Analysis Two radiologists with 11 and 8 years of experience with thyroid ultrasound and FNA performed all sonography examinations and FNA procedures. All sonography examinations were performed using an HDI 5000 (ATL Ultrasound) or an IU 22 (Philips Healthcare) instrument equipped with either a 7-15 or a 5-12 MHz linear-array transducer. Sonography characteristics and sonography diagnoses were prospectively recorded and classified according to the standardized lexicon developed by Moon et al. [3]. The internal content of the nodule was classified as solid, predominantly solid, predominantly cystic, or cystic. Nodule shape was classified as ovoid to round, irregular, or taller than wide, and nodule margins were categorized as well-defined smooth, ill-defined, or spiculated. The echogenicity of the nodule was compared with that of thyroid parenchyma and was classified as hyperechoic, isoechoic, hypoechoic, or marked hypoechogenicity; the latter designation was used when the nodule was hypoechoic compared with adjacent strap muscle. Calcification, if present, was classified as microcalcification, macrocalcification, or rim calcification. Reviewers also assessed whether the nodule appeared spongiform and whether the comet-tail artifact was present. Final assessments were categorized as suspicious for malignancy, indeterminate, or probably benign. Findings suspicious for malignancy on sonography were defined as having at least one of the following features: a taller-than-wide shape, a spiculated margin, marked hypoechogenicity, or micro- or macrocalcifications [3]. Probably benign findings were defined as the presence of a spongiform nodule or of a predominantly cystic or cystic mass with comet-tail artifact. Indeterminate findings were defined as all sonography findings other than those indicating probable benignity or a suspicious malignant nodule. The size of the nodules was considered the maximum diameter as measured on sonography. Cytologic Analysis When a single thyroid nodule was found on sonography, FNA of the suspicious nodule was performed according to the sonographic assessment regardless of the size of the nodule. For a nodule smaller than 5 mm, FNA was performed according to patient s risk factors and the feasibility of FNA. In case of multiple thyroid nodules in a patient, FNA was performed for the nodule that appeared most suspicious. When multiple nodules in the same patient showed similar sonographic findings, the largest one was aspirated. Although a patient underwent multiple FNAs of multiple lesions, the sample for BRAF mutation analysis was obtained only for the nodule that appeared most suspicious on each side. FNAs were performed by the same radiologist who performed the sonographic examination using a freehand technique with a 22-gauge needle and a 5-mL disposable plastic syringe. Each nodule was aspirated at least twice. Cytology slides were reviewed to confirm the cytologic diagnosis by an experienced pathologist. Based on preoperative cytologic diagnoses, FNA cytology results were categorized as malignant, benign, indeterminate, or nondiagnostic. The indeterminate category encompassed samples showing hypercellularity suggestive of a follicular neoplasm (including the follicular variant of PTC) and samples showing atypical cytoarchitectural features suggestive of, but not diagnostic for, malignancy. No case had a nondiagnostic cytologic result by preoperative FNA. BRAF Mutation Analysis Cytology slides were retrieved for BRAF V600E analysis. After removing coverslips from FNA cytology slides, atypical cells of interest were removed and genomic DNA was extracted. Genomic DNA (5 μl) was amplified by polymerase chain reaction (PCR) using standard conditions [13]. The SNU-790 human papillary carcinoma cell line containing heterozygous BRAF V600E was used as a control for BRAF mutant, as previously reported [17], and the SNU-216 human stomach cancer cell line without BRAF V600E was used as a wild-type control. Each PCR mix included forward and reverse primers (0.4 pmol each) and 5 μl of genomic DNA. Biotinylated PCR product (20 μl) was attached to streptavidin-sepharose beads according to a standard protocol. The PCR primer and sequencing primer sequences were as follows: exon 15(T1799A) forward, 5 -biotin-cttcata- ATGCTTGCTCTGATAGG-3 ; reverse, 5 -GGC- CAAAAATTTAATCAGTGGAA-3 ; and sequencing reverse, 5 -CACTCCATCGAGATT T-3. Pyrosequencing was performed using a single-nucleotide polymorphism reagent kit and a sample preparation tool (PSQ 96, Biotage). Statistical Analysis To compare BRAF V600E -positive and BRAF V600E - negative PTC patients, the chi-square test and Fisher exact test were used to analyze categoric variables (i.e., clinicopathologic findings and sonographic findings), and the Student t test and Mann-Whitney test were used to analyze numeric variables. In addition, multiple logistic regression analysis with the forward stepwise method was performed to identify independent sonography predictors of BRAF status. The McNemar chi-square test was used to compare the diagnostic sensitivities and specificities of sonography, cytology, BRAF status, sonography plus cytology, and BRAF status plus cytology for the diagnosis of malignancies in all 553 nodules. Multiple logistic regression analysis using the enter method was performed to identify independent predictors of malignancy from among age, sex, sonography features, cytology result, and AJR:198, March

3 Moon et al. TABLE 1: Clinicopathologic Characteristics of the 164 Patients With Papillary Thyroid Carcinoma (PTC) or the Follicular Variant of PTC by BRAF v600e Status BRAF v600e Feature BRAF status. To determine diagnostic sensitivities and specificities and to perform multiple regression analysis, sonography and cytology assessments were dichotomized according to the presence of malignant findings. When two diagnostic tests were combined, a nodule was considered positive for malignancy when either test was positive. Statistical significance was accepted for p values of < 0.05, and SPSS software (version 15.0, SPSS) for Microsoft Windows was used for the statistical analysis. Results BRAF V600E was detected in 141 of 170 malignant thyroid nodules (82.9%). Of these, 140 were conventional PTCs and one was a follicular-variant PTC. None of the 383 benign thyroid nodules harbored BRAF V600E. Clinicopathologic and Sonography Features of Papillary Thyroid Carcinomas by BRAF V600E Status The clinicopathologic characteristics of the PTCs (n = 164; 161 conventional PTCs and three follicular-variant PTCs) by BRAF V600E status are summarized in Table 1. Although BRAF V600E showed a tendency toward extracapsular invasion (p = 0.072), it was not significantly associated with patient age, tumor size, tumor multifocality, lymph node metastasis, or tumor stage. Positive (n = 141) Negative (n = 23) Age (y) Mean ± SD 47.0 ± ± Sex No. (%) of men 31 (22.0) 4 (17.4) Tumor size (mm) Mean ± SD 11.8 ± ± No. (%) of nodules 10 mm 93 (66.0) 14 (60.9) No. (%) of nodules > 10 mm 48 (34.0) 9 (39.1) Multifocality, no. (%) of nodules 30 (21.3) 5 (21.7) 1 Extracapsular invasion, no. (%) of nodules 65 (46.1) 6 (26.1) Lymph node metastasis, no. (%) of nodules Central 44 (31.2) 6 (26.1) Lateral 6 (4.3) 0 (0.0) Tumor stage, no. (%) of nodules I or II 90 (63.8) 14 (60.9) III or IV 51 (36.2) 9 (39.1) No significant differences were found between BRAF V600E -positive PTCs and BRAF V600E -negative PTCs with respect to sonography-determined contents, margins, echogenicities, or calcifications (Table 2 and Figs. 1 and 2). Only an irregular shape was found to have a negative association with BRAF V600E status by Fisher exact test and multiple logistic regression analysis (p = 0.004). Of the 164 PTC nodules, 158 (96.3%) were classified as suspicious for malignancy by sonography. Only six nodules (3.7%) were classified as indeterminate by sonography, and no significant difference was found between BRAF V600E -positive PTCs and BRAF V600E -negative PTCs in terms of the frequencies of malignant sonography features. Relationship Between Sonography Findings and BRAF V600E Status for Papillary Thyroid Carcinomas With an Indeterminate Cytology Finding Of all 164 PTC nodules, none was diagnosed as benign on cytology. An indeterminate cytology finding (n = 29) was more frequent for BRAF V600E -negative PTCs (34.8%, n = 8/23) than for BRAF V600E -positive PTCs (14.9%, n = 21/141) (p = 0.035). Of the 29 PTCs with an indeterminate cytology result, sonographic findings were malignant in 27 (93.1%) and benign in two (6.9%). For PTCs with an indeterminate cytology result, BRAF V600E status was not found to be related to diagnosis by sonography (p = 0.483) (Table 3). Comparisons of the Diagnostic Values of Sonography, Cytology, BRAF V600E Status, and Their Combinations The sensitivity and specificity of sonography for diagnosing malignant thyroid nodules were 94.1% and 93.2%, respectively (Table 4). When sonography and cytology were considered in combination, they showed significantly higher sensitivity than sonography alone (97.6% vs 94.1%, respectively; p = 0.014) but no significant difference in terms of specificity than sonography alone (p = 1.0). The specificity of BRAF V600E status was 100%. By adding BRAF V600E analysis to cytology, sensitivity for malignancy significantly increased (94.1%) as compared with cytology alone (81.8%) (p < 0.001). When cytology was combined with sonography or BRAF V600E status, sonography plus cytology showed a tendency to have a higher sensitivity than cytology plus BRAF V600E (p = 0.06). Finally, the triple combination of sonography, cytology, and BRAF V600E status showed higher sensitivity than BRAF V600E and cytology (98.2% vs 94.1%, respectively) (p = 0.023). On the other hand, multiple regression analysis showed that age, sex, sonography status, p 670 AJR:198, March 2012

4 BRAF Analysis and Sonography to Diagnose PTC TABLE 2: Sonographic Characteristics of 164 Patients With Papillary Thyroid Carcinoma by BRAF v600e Status BRAF v600e Feature and cytology status, but not BRAF V600E status, were independent predictors of the presence of a thyroid malignancy (Table 5). Discussion We found that BRAF V600E status in patients with PTC is not associated with sonography features with the exception of a negative relation with an irregular shape. In addition, our results showed that combining BRAF V600E mutation status and sonographic findings with FNA cytology findings enhanced diagnostic performance by increasing sensitivity. Various oncogenic mutations have been reported in thyroid cancer [18 20]. Of these, the activating point mutation in BRAF is the most common in PTC [21]. BRAF is the most potent of the three isoforms of rapidly accelerated fibrosarcoma (RAF) kinase and is a serine threonine kinase that participates in the mitogenactivated protein kinase pathway [22, 23]. This pathway normally contributes to cell proliferation, cell differentiation, apoptosis, and cell survival [24, 25], and when it is aberrantly activated, tumorigenesis ensues [22, 26 28]. Positive (n = 141) Negative (n = 23) Content Solid 130 (92.2) 23 (100) Predominantly solid 7 (5.0) 0 (0) 1 Predominantly cystic 4 (2.8) 0 (0) 1 Shape Ovoid to round 90 (63.8) 12 (52.2) Irregular 7 (5.0) 6 (26.1) Taller than wide 44 (31.2) 5 (21.7) Margin Well-defined smooth 13 (9.2) 3 (13.0) Ill-defined 43 (30.5) 9 (39.1) Spiculated 85 (60.3) 11 (47.8) Echogenicity Hyperechogenic 2 (1.4) 0 (0) 1 Isoechoic 0 (0) 1 (4.3) Hypoechoic 50 (35.5) 9 (39.1) Marked hypoechoic 89 (63.1) 13 (56.5) Calcification Microcalcification 96 (68.1) 18 (78.3) Macrocalcification 8 (5.7) 1 (4.3) 1 Presence of any suspicious sonographic finding 137 (97.2) 21 (91.3) Note Except p values, data are presented as no. (%) of nodules. Although a meta-analysis and multicenter study revealed that BRAF V600E is significantly associated with extrathyroidal extension, a higher clinical stage, and lymph node metastasis [6, 29], recent studies have produced conflicting results regarding the relation between BRAF V600E status and a poor prognosis. Specifically, BRAF V600E in micro- PTC (n = 339) was found to be associated with tumor size, extracapsular invasion, and a high TNM stage but not with lymph node metastasis [12]. Hwang et al. [15] reported that BRAF V600E mutation status in 135 PTCs was not related to extrathyroidal extension or lymph node metastases, and a Japanese study of 631 PTCs found no relationship between BRAF V600E status and any high-risk factor such as lymph node metastasis, extrathyroid extension, advanced age, and distal metastasis [30]. In our study, extracapsular invasion was found to be related to the presence of BRAF V600E, as has been reported previously [6, 7, 29]. However, lymph node metastasis was not found to be associated with the BRAF mutation in the current study. Our result is consistent with those of several Korean and Japanese studies [12, 15, 30]. The lack of an association between the BRAF mutation and several known prognostic factors in our study could be a result of a higher frequency of microcarcinoma or a lower prevalence of the tall cell variant, which is known to adopt an aggressive course [31]. To our knowledge, only two recent reports have assessed the relationship between sonography features and BRAF V600E status in patients with PTC [12, 15]. Our results support those of previous studies that found that most sonography features are not related to BRAF V600E status. Only marked hypoechogenicity [12] and calcification [15] have been reported to be associated with BRAF V600E status in previous studies. In contrast, we found that only an irregular shape was significantly associated with the absence of BRAF V600E. In a previous study, an irregular shape was not found to be a specific finding for either malignancy or benignancy [3]. Our findings imply that BRAF V600E status might affect tumor growth patterns and that this association might be responsible for the differ- p AJR:198, March

5 Moon et al. Fig. 1 Papillary thyroid carcinoma in 42-year-old woman. Transverse sonography image shows 7-mm marked hypoechoic nodule (cursors) with taller-than-wide shape in right thyroid gland that was classified as being malignant on basis of sonography features. Lesion was positive for V600E mutation of BRAF. Indeterminate cytology result was obtained by fine-needle aspiration but papillary thyroid carcinoma was proven by surgery. ent shapes observed at sonography. However, this finding seems to be applicable primarily to microcarcinomas and not to all PTCs because the majority of PTCs in our study (69%) were 10 mm or smaller. The usefulness of BRAF V600E analysis in nodules with an indeterminate FNA cytology result has been previously reported [11, 13, 14, 32]. The BRAF V600E mutation is believed to be specific for PTC because no report has been issued of this mutation in benign nodules [33]. An indeterminate cytology result accounts for 15 25% of FNA results of thyroid nodules [5] and carries a 20 30% risk of malignancy [16]. As compared with BRAF V600E - positive PTC, the higher frequency of indeterminate cytology in BRAF V600E -negative PTC in our study implies that BRAF mutation analysis on an FNA specimen may not be as helpful as indicated by previous studies [11, 13, 14, 32]. In contrast to a previous study [11], we found that BRAF V600E was present in sonography-positive and sonography-negative nodules without a significance difference. Considering the fact that 74.1% of nodules (20/27) with a malignant sonography finding and an indeterminate cytology result were found to harbor BRAF V600E, BRAF mutation analysis appears to be justifiable to confirm malignancy if a nodule has suspicious sonography findings and an indeterminate cytology result. Despite a recent surge of studies on BRAF V600E analysis in relation to FNA cytology results and sonography features, only two reports have compared the diagnostic accuracies of cytology and BRAF V600E status and combinations of the two [11, 13]. BRAF V600E status alone has a specificity of 100% and a positive predictive value of 100% for diagnosing malignancy in PTC [13]. Nam et al. [11] reported significant improvements in the sensitivity, negative predictive value, and diagnostic accuracy of FNA cytology for diagnosing malignancy by adding BRAF V600E analysis. Furthermore, although sonography has been reported to be an adjunctive diagnostic test or guidance tool for FNA [34], recent studies support the expansion of the role of sonography for diagnosing malignancy in thyroid nodules [3, 35, 36]. We found that either sonography features or BRAF V600E status can increase the diagnostic sensitivity of diagnosis when combined Fig. 2 Papillary thyroid carcinoma in 30-year-old woman. Transverse sonography image shows irregularly shaped and marked hypoechoic nodule that contains microcalcifications (arrows). Lesion was negative for V600E mutation of BRAF. Papillary thyroid carcinoma was proven by fine-needle aspiration and surgery. TABLE 3: Sonographic Characteristics and BRAF Mutation Status in Patients With an Indeterminate Cytology Result Sonographic Finding BRAF v600e Positive (n = 21) Negative (n = 8) Indeterminate 1 (4.8) 1 (12.5) Malignant 20 (95.2) 7 (87.5) Note Except p values, data are presented as no. (%) of nodules. with FNA cytology results. Furthermore, the combination of sonography, cytology, and BRAF V600E status was found to be superior to the BRAF V600E status cytology combination in terms of sensitivity. Thus, the integration of sonography and cytology findings is likely to aid diagnosis in clinical practice. Our results also suggest that a combination of sonography and cytology can provide an alternative diagnostic pathway to cytology plus BRAF V600E molecular analysis when the genetic analysis is not available. Accordingly, we suggest that BRAF V600E analysis be reserved for specific situations, such as for risk stratification or for solution against equivocal cytology [12, 37]. Our suggestion is also supported by the fact that the results of our multiple regression analyses do not agree with the results of McNemar analysis. In view of these conflicting results together, we cannot recommend that BRAF mutation analysis be conducted as a routine additional test for the diagnosis of thyroid malignancy but rather that it be used as an additional test when sonography and cytology results disagree. A few limitations of the current study should be addressed. First, the prevalence of malignan- p 672 AJR:198, March 2012

6 BRAF Analysis and Sonography to Diagnose PTC TABLE 4: Diagnostic Performances of Sonography, Aspiration Cytology, and BRAF v600e Mutation Analysis Test Sensitivity Specificity Accuracy Sonography % (95% CI) 94.1 ( ) 93.2 ( ) 93.5 ( ) Raw data (no. of relevant cases / total no.) 160/ / /553 Cytology % (95% CI) 81.8 ( ) 99.5 ( ) 94.0 ( ) Raw data (no. of relevant cases / total no.) 139/ / /553 BRAF mutation status % (95% CI) 82.9 ( ) ( ) 94.8 ( ) Raw data (no. of relevant cases / total no.) 141/ / /553 Sonography and cytology % (95% CI) 97.6 ( ) 93.2 ( ) 94.6 ( ) Raw data (no. of relevant cases / total no.) 166/ / /553 Cytology and BRAF mutation status % (95% CI) 94.1 ( ) 99.5 ( ) 97.8 ( ) Raw data (no. of relevant cases / total no.) 160/ / /553 Sonography, cytology, and BRAF mutation status % (95% CI) 98.2 ( ) 93.2 ( ) 94.8 ( ) Raw data (no. of relevant cases / total no.) 167/ / /553 TABLE 5: Multiple Logistic Regression Analysis for Independent Factors for Predicting Thyroid Malignancy Factor β-coefficient Odds Ratio 95% CI p Cytology ± < Sonography ± < Age ± Sex ± BRAF mutation status ± cies was relatively high in our study. This may have been due, in part, to the retrospective design in our study. In addition, most suspicious nodules were analyzed for BRAF mutation status, which probably increased the prevalence of malignancies and affected diagnostic accuracy. Second, many nodules in our study were smaller than 1 cm and may not have been biopsied in a different situation. According to a recent guideline [34], the threshold size for aspiration is 5 mm for a nodule with malignant sonography findings. However, the fate and prognosis of microcarcinomas remain subjects of debate; for example, even micro-ptc exhibits substantial lymph node metastasis (8 50%) and recurrence rate (1 7%) [38]. In Korea, the recommendation is that FNA be performed if feasible on any nodule suspected of being malignant on the basis of sonography findings and any nodule larger than 5 mm [39]. For nodules smaller than 5 mm, selective FNA can be performed depending on patient risk factors and radiologist experience [39]. Accordingly, the sensitivity and specificity of sonography, cytology, and BRAF mutation analysis in our study may not represent those of studies in other countries. Third, although both radiologists were experienced, interobserver variability might have affected sonography findings. Finally, the interpretation of cytology results might also have been affected by intraobserver variability. In conclusion, adding sonography and BRAF mutation analysis to FNA cytology would increase the sensitivity of preoperative diagnosis for malignancy in PTC. Acknowledgment This study was supported by Konkuk University. References 1. Papini E, Guglielmi R, Bianchini A, et al. Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-doppler features. J Clin Endocrinol Metab 2002; 87: Kim EK, Park CS, Chung WY, et al. New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. AJR 2002; 178: Moon WJ, Jung SL, Lee JH, et al. Benign and malignant thyroid nodules: US differentiation multicenter retrospective study. Radiology 2008; 247: Gharib H, Goellner JR, Zinsmeister AR, Grant CS, Van Heerden JA. Fine-needle aspiration biopsy of the thyroid: the problem of suspicious cytologic findings. Ann Intern Med 1984; 101: Alexander EK. Approach to the patient with a cytologically indeterminate thyroid nodule. J Clin Endocrinol Metab 2008; 93: Xing M. BRAF mutation in thyroid cancer. Endocr Relat Cancer 2005; 12: Xing M. BRAF mutation in papillary thyroid cancer: pathogenic role, molecular bases, and clinical implications. Endocr Rev 2007; 28: Kim KH, Kang DW, Kim SH, Seong IO, Kang AJR:198, March

7 Moon et al. DY. Mutations of the BRAF gene in papillary thy- ic acid and mutations in the BRAF gene. Mol Cell 31. Michels JJ, Jacques M, Henry-Amar M, Bardet S. roid carcinoma in a Korean population. Yonsei Endocrinol 2007; 264: Prevalence and prognostic significance of tall cell Med J 2004; 45: Nikiforov YE. RET/PTC rearrangement in thy- variant of papillary thyroid carcinoma. Hum 9. Kim TY, Kim WB, Rhee YS, et al. The BRAF roid tumors. Endocr Pathol 2002; 13:3 16 Pathol 2007; 38: mutation is useful for prediction of clinical recurrence in low-risk patients with conventional papillary thyroid carcinoma. Clin Endocrinol (Oxf) 2006; 65: Chung KW, Yang SK, Lee GK, et al. Detection of BRAF V600E mutation on fine needle aspiration specimens of thyroid nodule refines cyto-pathology diagnosis, especially in BRAF 600E mutation prevalent area. Clin Endocrinol (Oxf) 2006; 65: Nam SY, Han BK, Ko EY, et al. BRAF V600E mutation analysis of thyroid nodules needle aspirates in relation to their ultrasonographic classification: a potential guide for selection of samples for molecular analysis. Thyroid 2010; 20: Kwak JY, Kim EK, Chung WY, Moon HJ, Kim MJ, Choi JR. Association of BRAF V600E mutation with poor clinical prognostic factors and US features in Korean patients with papillary thyroid microcarcinoma. Radiology 2009; 253: Kim SK, Kim DL, Han HS, et al. Pyrosequencing analysis for detection of a BRAF V600E mutation in an FNAB specimen of thyroid nodules. Diagn Mol Pathol 2008; 17: Jo YS, Huang S, Kim YJ, et al. Diagnostic value of pyrosequencing for the BRAF V600E mutation in ultrasound-guided fine-needle aspiration biopsy samples of thyroid incidentalomas. Clin Endocrinol (Oxf) 2009; 70: Hwang J, Shin JH, Han BK, et al. Papillary thyroid carcinoma with BRAF V600E mutation: sonographic prediction. AJR 2010; 194:1344; [web] W425 W Baloch ZW, Cibas ES, Clark DP, et al. The National Cancer Institute thyroid fine needle aspiration state of the science conference: a summation. Cytojournal 2008; 5:6 17. Koh CS, Ku JL, Park SY, et al. Establishment and characterization of cell lines from three human thyroid carcinomas: responses to all-trans-retino- 19. Kroll TG, Sarraf P, Pecciarini L, et al. PAX8- PPARgamma1 fusion oncogene in human thyroid carcinoma [corrected]. Science 2000; 289: [Erratum in Science 2000; 289:1474] 20. Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature 2002; 417: Costa AM, Herrero A, Fresno MF, et al. BRAF mutation associated with other genetic events identifies a subset of aggressive papillary thyroid carcinoma. Clin Endocrinol (Oxf) 2008; 68: Mercer KE, Pritchard CA. Raf proteins and cancer: B-Raf is identified as a mutational target. Biochim Biophys Acta 2003; 1653: Wellbrock C, Karasarides M, Marais R. The RAF proteins take centre stage. Nat Rev Mol Cell Biol 2004; 5: MacCorkle RA, Tan TH. Mitogen-activated protein kinases in cell-cycle control. Cell Biochem Biophys 2005; 43: Robinson MJ, Cobb MH. Mitogen-activated protein kinase pathways. Curr Opin Cell Biol 1997; 9: Kohno M, Pouyssegur J. Targeting the ERK signaling pathway in cancer therapy. Ann Med 2006; 38: Sebolt-Leopold JS, Herrera R. Targeting the mitogen-activated protein kinase cascade to treat cancer. Nat Rev Cancer 2004; 4: Garnett MJ, Marais R. Guilty as charged: B-RAF is a human oncogene. Cancer Cell 2004; 6: Lee JH, Lee ES, Kim YS. Clinicopathologic significance of BRAF V600E mutation in papillary carcinomas of the thyroid: a meta-analysis. Cancer 2007; 110: Ito Y, Yoshida H, Maruo R, et al. BRAF mutation in papillary thyroid carcinoma in a Japanese population: its lack of correlation with high-risk clinicopathological features and disease-free survival of patients. Endocr J 2009; 56: Moon HJ, Kwak JY, Kim EK, et al. The role of BRAF V600E mutation and ultrasonography for the surgical management of a thyroid nodule suspicious for papillary thyroid carcinoma on cytology. Ann Surg Oncol 2009; 16: Pizzolanti G, Russo L, Richiusa P, et al. Fineneedle aspiration molecular analysis for the diagnosis of papillary thyroid carcinoma through BRAF V600E mutation and RET/PTC rearrangement. Thyroid 2007; 17: American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer; Cooper DS, Doherty GM, Haugen BR, et al. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2009; 19: [Errata in Thyroid 2010; 20: and 2010; 20:942] 35. Moon WJ, Kwag HJ, Na DG. Are there any specific ultrasound findings of nodular hyperplasia ( leave me alone lesion) to differentiate it from follicular adenoma? Acta Radiol 2009; 50: Horvath E, Majlis S, Rossi R, et al. An ultrasonogram reporting system for thyroid nodules stratifying cancer risk for clinical management. J Clin Endocrinol Metab 2009; 94: Kim SK, Hwang TS, Yoo YB, et al. Surgical results of thyroid nodules according to a management guideline based on the BRAF V600E mutation status. J Clin Endocrinol Metab 2011; 96: Pazaitou-Panayiotou K, Capezzone M, Pacini F. Clinical features and therapeutic implication of papillary thyroid microcarcinomas. Thyroid 2007; 17: Moon WJ, Baek JH, Jung SL, et al.; Korean Society of Thyroid Radiology; Korean Society of Radiology. Ultrasonography and the ultrasoundbased management of thyroid nodules: consensus statement and recommendations. Korean J Radiol 2011; 12: AJR:198, March 2012