Detection of BRAF c.1799T >A (p.V600E) Mutation Using Residual Routine Fine-Needle Aspiration Specimens of Papillary Thyroid Carcinoma Huan Zhao,1 Zhi-hui Zhang,1 Bin Zhou,1 Ting Xiao,2 Qin-jing Pan,1 and Hui-qin Guo1*
Background: BRAF p.V600E mutation is the most common
genetic alteration in papillary thyroid cancer (PTC) and has been used as a diagnostic and prognostic marker in PTC. The aim of this study was to investigate the utility of preoperative BRAF p.V600E mutation analysis as an adjunctive diagnostic and prognostic tool to routine fine-needle aspiration (FNA). Methods: Specimens were collected from thyroid nodules by FNA. Cytology diagnosis and BRAF p.V600E testing were performed on these specimens. Molecular and cytological results were correlated with histology outcomes. Results: A total of 195 patients with thyroid nodules were enrolled, including 25 benign lesions and 170 PTCs. BRAF p.V600E testing was successfully performed in all specimens. The combination of BRAF p.V600E testing and cytology improved the sensitivity of cytology from 70% to 85.3% (P 5 0.001). This significant increase in sensitivity was due to the detection of PTC by BRAF p.V600E testing in the nodules with atypical or suspicious PTC cytology results. Patients with BRAF p.V600E-positive tumors were significantly older than those who did not harbor mutations (45.6 years vs. 39.8 years, P 5 0.002). No correlations between BRAF p.V600E mutation and other clinical-pathology parameters were observed. 1 Department of Pathology, Cancer Institute/Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, People’s Republic of China 2 State Key Laboratory of Molecular Oncology, Department of Etiology and Carcinogenesis, Cancer Institute/Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, People’s Republic of China Contract grant sponsor: Youth Fund of Peking Union Medical College. Conflict of interest: The authors have no conflicts of interest to declare. *Correspondence to: H.-Q. Guo, Associate Prof., Department of Pathology, Cancer Institute/Hospital Chinese Academy of Medical Sciences, 17 Nanli Panjiayuan Lane, Chaoyang District, Beijing 100021, People’s Republic of China. E-mail: [email protected] Received 25 June 2014; Revised 10 April 2015; Accepted 19 June 2015 DOI: 10.1002/dc.23302 Published online 7 July 2015 in Wiley Online Library (wileyonlinelibrary.com).
786
Diagnostic Cytopathology, Vol. 43, No 10
Conclusions: Detection of BRAF p.V600E mutation can be successfully carried out using residual liquid-based materials. It can be performed as a diagnostic tool to supplement traditional thyroid FNA, especially in cases with atypical or suspicious PTC. However, the role of BRAF p.V600E in guidance of the extent of thyroidectomy and nodal clearance requires further C 2015 Wiley study. Diagn. Cytopathol. 2015;43:786–790. V Periodicals, Inc.
Key Words: papillary thyroid carcinoma; BRAF p.V600E mutation; fine-needle aspiration; diagnosis; prognosis
Fine-needle aspiration (FNA) is now the most successful diagnostic test in the management of thyroid nodules.1–3 However, FNA may also be nondiagnostic or may demonstrate indeterminate or suspicious cytologic features in 20–30% of all biopsies.4–7 Hence, an ancillary test for suspicious FNA results is required. Moreover, the detection rate of microcarcinoma is increasing with the widespread use of ultrasound. However, the optimal management of microcarcinoma remains controversial, and often focuses on the extent of thyroidectomy and lymph node clearance. Thus, additional prognostic information provided by preoperative FNA may be helpful for surgeons in making a decision on initial treatment. BRAF c.1799T > A (p.V600E) mutation is the most common genetic alteration in papillary thyroid cancer (PTC) and is found in 15–80% of these tumors and confined to malignant nodules.8–10 It is possible to speculate a diagnostic role of BRAF p.V600E testing in PTC. The prognostic significance of the BRAF p.V600E mutation has been explored in several studies, in which BRAF p.V600E-positive PTC was found to have a more aggressive behavior.8,11–18 In this study, we investigated the usefulness of preoperative BRAF p.V600E mutation analysis as an adjunctive diagnostic and prognostic tool to routine FNA. C 2015 WILEY PERIODICALS, INC. V
Diagnostic Cytopathology DOI 10.1002/dc
BRAF P.V600E TESTING IN THYROID FNA
Materials and Methods
DNA Extraction
The materials were obtained from the Cancer Institute/ Hospital, Chinese Academy of Medical Sciences (CAMS) between November 2010 and July 2012. The FNA samples were collected preoperatively and intraoperatively. Of the 195 cases, 95 samples were obtained in the outpatient clinic before surgery. The remaining 100 samples were obtained at the time of frozen examination. Intraoperative FNA biopsies were performed on surgically resected thyroid nodules by one of the two cytopathologists (Dr. Guo and Dr. Zhao). The thyroid nodules were cut in half, one half was sent for frozen pathologic diagnosis and the other half was used for FNA. FNA was performed under direct visualization. At least two passes were performed on each tumor. One pass aspirate was expressed onto a glass slide for cytologic diagnosis. The other pass was rinsed in PreservCyt solution (Hologic, Marlborough, MA, USA) for DNA extraction. FNA was repeated one or more times until suspended white particles were observed in the PreservCyt solution macroscopically. Preoperative FNA samples were retrospectively collected after obtaining clinical cytologic diagnoses. FNA biopsies were performed with one to two passes. Palpable lesions were aspirated by a cytopathologist in the FNA clinic; nonpalpable lesions using ultrasound-guided FNA were aspirated by radiologists. Aspirates were rinsed in PreservCyt solution. One ThinPrep (Hologic, Marlborough, MA, USA) slide was prepared from each case for assessment of both morphology and cellularity. Residues were collected for DNA extraction. In this way, we only selected the specimens which were diagnosed as PTC or suspicious PTC by cytology. Residue selection criteria were defined as 10 groups of cells on the slide in 10 ml of PreservCyt solution. Previous studies suggested selecting 6–10 groups of cells on the slide to perform PCR detection.19 In order to guarantee enough DNA, our residue selection was based on both cellularity and volume. Approximately 50% of routine FNA specimens met the selection criteria and were used for molecular testing. The liquid-based materials were stored at 48C and submitted for molecular testing within 3 months. This study was reviewed and approved by the ethics committee of the Cancer Institute/Hospital, CAMS. All patients gave informed consent.
Cells in PreservCyt solution were first centrifuged, then incubated in 500 ll of DNA lysis solution [1 mg/ml proteinase K, 10 mmol/L Tris–HCl (pH 8.0), 0.1 mol/L EDTA (pH 8.0), 0.5% (w/v) SDS] at 558C for approximately 12 hours, and the DNA was then extracted by phenol and chloroform, and stored at 2208C until use. The nucleic acids were quantified using a NanoDrop spectrophotometer.
Cytology Diagnosis According to the The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC), cytological diagnoses were categorized as benign, atypia of indeterminate significance (AUS), follicular neoplasm (FN), suspicious malignant/suspicious papillary thyroid carcinoma (SM/ SPTC), and malignant/PTC.
Detection of BRAF P.V600E Mutation The primer sets for PCR amplification were designed as follows: BRAF p.V600E (codon 600), 50 -TCATAATGCT TGCTCTGATAGGA-30 (forward), and 50 -GGCCAAAAA TTTAATCAGTGGA-30 (reverse). PCR was carried out using TaKaRa Taq DNA polymerase (TaKaRa, in a final volume of 50 ml, with 13 TaKaRa Taq Buffer with Mg21, 200 lM of each dNTP, 200 nM of forward and reverse primers, 1.25 U of TaKaRa Taq, and 50–100 ng of DNA template). PCR cycling parameters were as follows: an initial denaturation at 958C for 5 min followed by 35 cycles at 958C for 30 seconds, 55–618C for 45 seconds, 728C for 30 seconds; and a final extension at 728C for 10 min. Five microliters of PCR product was run on a 1.5% agarose gel to check the specificity of amplification, and 20 ml was sent for direct sequencing with forward and reverse primers for each gene (SinoGenoMax Co., Ltd., China).
Statistical Analysis The histological diagnosis in each patient was obtained from their medical records. The results of histopathology were taken as the gold standard. Cytological and molecular results were compared with histopathology. Group comparisons of categorical variables were performed using the v2 test. The association between clinicopathologic factors and presence of the BRAF p.V600E mutation was validated by the Pearson v2 and independent t tests. Statistical analysis was carried out using SPSS 13.0 and P < 0.05 was considered statistically significant.
Results A total of 195 FNA samples from 44 male and 151 female patients (mean age: 45.1 years), were included in this study. All FNA specimens were sufficient to successfully perform DNA extraction and PCR analysis. On surgical follow-up, 170 cases were confirmed as PTC, and the lesions in the other 25 cases were benign. Using cytology alone, the diagnostic sensitivity, specificity, and accuracy for the detection of PTC were 70% (119/170), 100% (25/25), and 73.8% (144/195), respectively. BRAF p.V600E testing showed similar sensitivity (67.1%), specificity (100%), and accuracy (71.3%) to that Diagnostic Cytopathology, Vol. 43, No 10
787
Diagnostic Cytopathology DOI 10.1002/dc
ZHAO ET AL. Table I. Diagnostic Values of FNA Cytology, BRAF p.V600E Testing or a Combination of Both for Detecting PTC Diagnostic modality Cytologya Cytologyb BRAF p.V600E testing Combineda Combinedb
SN (%)
SP (%)
PPV (%)
NPV (%)
AC (%)
70.0 91.8 67.1
100.0 84.0 100.0
100.0 97.5 100.0
32.9 60.0 30.9
73.8 90.8 71.3
85.3 92.9
100.0 84.0
100.0 97.5
50.0 63.6
87.2 91.8
a
Cytology diagnosis of malignant was taken as the diagnostic threshold. b Cytology diagnosis of suspicious malignant was taken as the diagnostic threshold. Abbreviations: SN, sensitivity; SP, specificity; PPV, positive predictive value; NPV, negative predictive value; AC, accuracy. Table II. Correlation Between Cytologic Diagnoses and BRAF p.V600E Mutation Status
Cytologic diagnoses (n) PTC (n 5 170)
Total Benign (n 5 25)
Total
PTC SPTC FN AUS Benign SPTC FN AUS Benign
119 37 2 7 5 170 4 2 1 18 25
BRAF p.V600E (1) (n)
(%)
88 24 0 2 0 114 0 0 0 0 0
73.9 64.9 0 28.6 0 67.1 0 0 0 0 0
Abbreviations: AUS, atypia of undetermined significance; FN, follicular neoplasm; SPTC, suspicious papillary thyroid carcinoma; PTC: papillary thyroid carcinoma.
of cytology. When FNA cytology was combined with BRAF p.V600E testing, improved sensitivity (85.3% vs 70.0%, P 5 0.001) and accuracy (87.2% vs 73.8%, P 5 0.001) were obtained for PTC diagnosis (Table I). Table II shows the correlation between BRAF p.V600E mutations and cytological diagnoses. In terms of the two cytology categories of SPTC and AUS, BRAF p.V600E testing contributed to cytology diagnosis. In the group with SPTC (n 5 41), 24 samples had BRAF p.V600E mutations. At final histology, 37 were confirmed as PTC by histopathology and all BRAF p.V600E mutated cases were histologically confirmed as PTC. BRAF p.V600E testing confirmed 64.9% (24/37) of PTC with SPTC cytology diagnosis. In the group with AUS cytology (n 5 8), 7 were PTC. Two samples had BRAF p.V600E mutations, and both were confirmed as PTC. BRAF p.V600E testing detected 28.6% (2/7) of PTC in AUS cytology samples. The clinical pathologic features of PTC according to BRAF p.V600E mutation status are shown in Table III. Of 170 PTCs, 13 recurrent patients (patients who already had a diagnosis of papillary carcinoma) were excluded from this 788
Diagnostic Cytopathology, Vol. 43, No 10
Table III. Univariate Analysis of the Association Between BRAF p.V600E Mutation and Clinicopathologic Factors BRAF p.V600E Positive (n 5 105) Characteristic
n
%
Age at diagnosis (years) Mean 6 SD 45.60 6 10.89 Range 21–76 Gender Male 23 21.9 Female 82 Capsular invasion Present 83 79 Absent 22 Extra-thyroid extension Present 49 46.7 Absent 56 Lymph node metastasis Present 54 51.5 Absent 51 Tumor size 1 cm 34 32.4 >1 cm 71 Multifocality Yes 41 39.0 No 64 AJCC stages I/II 78 74.3 III/IV 27
BRAF p.V600E Negative (n 5 52) n
%
39.8 6 10.80 18–67
P 0.002*
11 41
21.2
0.914
41 11
78.8
0.977
22 30
42.3
0.606
30 22
57.7
0.459
16 36
30.8
0.838
18 34
34.6
0.589
42 10
80.8
0.368
study. Of the other 157 cases, 123 were female and 34 were male aged 18–76 years, with an average age of 44 years. Patients with BRAF p.V600E positive tumors were significantly older than those who did not harbor the mutation (45.6 years vs. 39.8 years, P 5 0.002). However, BRAF p.V600E mutation was not associated with patient gender, capsular invasion, extra-thyroid extension, lymph node metastasis, tumor size, multifocality, and TNM stage. In addition, no significant relationships between BRAF p.V600E mutation and clinicopathologic characteristics were observed in the PTC group with tumor size 1 cm (Table IV).
Discussion It has been reported that BRAF p.V600E gene mutations can be successfully identified using liquid-based material with high reproducibility.20 Therefore, it is necessary to establish criteria regarding adequate cell number for molecular testing and to avoid false-negative findings. Previous studies suggested selecting 6–10 groups of cells on the slide to perform PCR detection.19 In order to guarantee enough DNA for analysis, we chose 10 groups of cells as the criterion in this study. With regard to the correct volume of tumor sample, most studies do not mention this. In this study, we were able to extract 300 ng to [mt]110,000 ng DNA from 10 ml of residual routine FNA specimens. This was sufficient to perform PCR detection. As this was an initial study on liquid-based residual specimens, further studies should be performed to
Diagnostic Cytopathology DOI 10.1002/dc
BRAF P.V600E TESTING IN THYROID FNA Table IV. Univariate Analysis of the Association Between BRAF p.V600E Mutation and Clinicopathologic Factors of PTCs with Tumor Size 1 cm BRAF p.V600E Positive (n 5 34) Characteristic
n
%
Age at diagnosis (years) Mean 6 SD 47.44 6 10.56 Range 25–65 Gender Male 6 17.6 Female 28 Capsular invasion Present 15 44.1 Absent 19 Extra-thyroid extension Present 11 32.4 Absent 23 Lymph node metastasis Present 12 35.3 Absent 22 Multifocality Yes 8 23.5 No 26 AJCC stages I/II 26 76.5 III/IV 8
BRAF p.V600E Negative (n 5 16) n
%
41.69 6 9.36 28–56
P 0.069
3 13
18.8
0.925
7 9
43.8
0.981
3 13
18.8
0.318
5 11
31.3
0.778
5 11
31.3
0.562
14 2
87.5
0.363
determine the smallest residual sample which can be used for molecular detection. BRAF p.V600E mutation is known to be highly specific to PTC and has not been reported in follicular carcinoma, adenoma, or benign thyroid nodules.21 Due to its high specificity, this genetic alteration is now considered a useful diagnostic marker for thyroid FNA. Zatelli et al. reported that BRAF p.V600E mutation analysis increased the sensitivity of cytology for PTC from 77% to 87%.22 Another author observed that the rate of PTC diagnosis using cytology alone was 62.3% (56/90), while the rate of PTC diagnosis using a morphomolecular method was 82.2% (74/90), with an increase of 20% in diagnostic accuracy due to BRAF p.V600E mutation analysis.23 In our study, we found that BRAF p.V600E mutation testing increased the sensitivity of the FNA procedure from 70% to 85.3%. Therefore, the addition of molecular analysis resulted in an increase in sensitivity of 15.3% compared to cytology alone. Diagnostic accuracy also increased from 73.8% to 87.2%. These findings are further evidence of the efficacy of BRAF p.V600E as a diagnostic marker in thyroid FNA. In the study by Xing, the authors believed that if BRAF p.V600E testing was performed before cytology examination, nearly half of the patients with PTC did not need to undergo cytology examination, which could result in a substantial cost saving.24 However, in clinical practice, the majority (60–80%) of FNA specimens are benign lesions and the risk of malignancy in cytology benign lesions is estimated to be
Background: BRAF p.V600E mutation is the most common
genetic alteration in papillary thyroid cancer (PTC) and has been used as a diagnostic and prognostic marker in PTC. The aim of this study was to investigate the utility of preoperative BRAF p.V600E mutation analysis as an adjunctive diagnostic and prognostic tool to routine fine-needle aspiration (FNA). Methods: Specimens were collected from thyroid nodules by FNA. Cytology diagnosis and BRAF p.V600E testing were performed on these specimens. Molecular and cytological results were correlated with histology outcomes. Results: A total of 195 patients with thyroid nodules were enrolled, including 25 benign lesions and 170 PTCs. BRAF p.V600E testing was successfully performed in all specimens. The combination of BRAF p.V600E testing and cytology improved the sensitivity of cytology from 70% to 85.3% (P 5 0.001). This significant increase in sensitivity was due to the detection of PTC by BRAF p.V600E testing in the nodules with atypical or suspicious PTC cytology results. Patients with BRAF p.V600E-positive tumors were significantly older than those who did not harbor mutations (45.6 years vs. 39.8 years, P 5 0.002). No correlations between BRAF p.V600E mutation and other clinical-pathology parameters were observed. 1 Department of Pathology, Cancer Institute/Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, People’s Republic of China 2 State Key Laboratory of Molecular Oncology, Department of Etiology and Carcinogenesis, Cancer Institute/Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, People’s Republic of China Contract grant sponsor: Youth Fund of Peking Union Medical College. Conflict of interest: The authors have no conflicts of interest to declare. *Correspondence to: H.-Q. Guo, Associate Prof., Department of Pathology, Cancer Institute/Hospital Chinese Academy of Medical Sciences, 17 Nanli Panjiayuan Lane, Chaoyang District, Beijing 100021, People’s Republic of China. E-mail: [email protected] Received 25 June 2014; Revised 10 April 2015; Accepted 19 June 2015 DOI: 10.1002/dc.23302 Published online 7 July 2015 in Wiley Online Library (wileyonlinelibrary.com).
786
Diagnostic Cytopathology, Vol. 43, No 10
Conclusions: Detection of BRAF p.V600E mutation can be successfully carried out using residual liquid-based materials. It can be performed as a diagnostic tool to supplement traditional thyroid FNA, especially in cases with atypical or suspicious PTC. However, the role of BRAF p.V600E in guidance of the extent of thyroidectomy and nodal clearance requires further C 2015 Wiley study. Diagn. Cytopathol. 2015;43:786–790. V Periodicals, Inc.
Key Words: papillary thyroid carcinoma; BRAF p.V600E mutation; fine-needle aspiration; diagnosis; prognosis
Fine-needle aspiration (FNA) is now the most successful diagnostic test in the management of thyroid nodules.1–3 However, FNA may also be nondiagnostic or may demonstrate indeterminate or suspicious cytologic features in 20–30% of all biopsies.4–7 Hence, an ancillary test for suspicious FNA results is required. Moreover, the detection rate of microcarcinoma is increasing with the widespread use of ultrasound. However, the optimal management of microcarcinoma remains controversial, and often focuses on the extent of thyroidectomy and lymph node clearance. Thus, additional prognostic information provided by preoperative FNA may be helpful for surgeons in making a decision on initial treatment. BRAF c.1799T > A (p.V600E) mutation is the most common genetic alteration in papillary thyroid cancer (PTC) and is found in 15–80% of these tumors and confined to malignant nodules.8–10 It is possible to speculate a diagnostic role of BRAF p.V600E testing in PTC. The prognostic significance of the BRAF p.V600E mutation has been explored in several studies, in which BRAF p.V600E-positive PTC was found to have a more aggressive behavior.8,11–18 In this study, we investigated the usefulness of preoperative BRAF p.V600E mutation analysis as an adjunctive diagnostic and prognostic tool to routine FNA. C 2015 WILEY PERIODICALS, INC. V
Diagnostic Cytopathology DOI 10.1002/dc
BRAF P.V600E TESTING IN THYROID FNA
Materials and Methods
DNA Extraction
The materials were obtained from the Cancer Institute/ Hospital, Chinese Academy of Medical Sciences (CAMS) between November 2010 and July 2012. The FNA samples were collected preoperatively and intraoperatively. Of the 195 cases, 95 samples were obtained in the outpatient clinic before surgery. The remaining 100 samples were obtained at the time of frozen examination. Intraoperative FNA biopsies were performed on surgically resected thyroid nodules by one of the two cytopathologists (Dr. Guo and Dr. Zhao). The thyroid nodules were cut in half, one half was sent for frozen pathologic diagnosis and the other half was used for FNA. FNA was performed under direct visualization. At least two passes were performed on each tumor. One pass aspirate was expressed onto a glass slide for cytologic diagnosis. The other pass was rinsed in PreservCyt solution (Hologic, Marlborough, MA, USA) for DNA extraction. FNA was repeated one or more times until suspended white particles were observed in the PreservCyt solution macroscopically. Preoperative FNA samples were retrospectively collected after obtaining clinical cytologic diagnoses. FNA biopsies were performed with one to two passes. Palpable lesions were aspirated by a cytopathologist in the FNA clinic; nonpalpable lesions using ultrasound-guided FNA were aspirated by radiologists. Aspirates were rinsed in PreservCyt solution. One ThinPrep (Hologic, Marlborough, MA, USA) slide was prepared from each case for assessment of both morphology and cellularity. Residues were collected for DNA extraction. In this way, we only selected the specimens which were diagnosed as PTC or suspicious PTC by cytology. Residue selection criteria were defined as 10 groups of cells on the slide in 10 ml of PreservCyt solution. Previous studies suggested selecting 6–10 groups of cells on the slide to perform PCR detection.19 In order to guarantee enough DNA, our residue selection was based on both cellularity and volume. Approximately 50% of routine FNA specimens met the selection criteria and were used for molecular testing. The liquid-based materials were stored at 48C and submitted for molecular testing within 3 months. This study was reviewed and approved by the ethics committee of the Cancer Institute/Hospital, CAMS. All patients gave informed consent.
Cells in PreservCyt solution were first centrifuged, then incubated in 500 ll of DNA lysis solution [1 mg/ml proteinase K, 10 mmol/L Tris–HCl (pH 8.0), 0.1 mol/L EDTA (pH 8.0), 0.5% (w/v) SDS] at 558C for approximately 12 hours, and the DNA was then extracted by phenol and chloroform, and stored at 2208C until use. The nucleic acids were quantified using a NanoDrop spectrophotometer.
Cytology Diagnosis According to the The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC), cytological diagnoses were categorized as benign, atypia of indeterminate significance (AUS), follicular neoplasm (FN), suspicious malignant/suspicious papillary thyroid carcinoma (SM/ SPTC), and malignant/PTC.
Detection of BRAF P.V600E Mutation The primer sets for PCR amplification were designed as follows: BRAF p.V600E (codon 600), 50 -TCATAATGCT TGCTCTGATAGGA-30 (forward), and 50 -GGCCAAAAA TTTAATCAGTGGA-30 (reverse). PCR was carried out using TaKaRa Taq DNA polymerase (TaKaRa, in a final volume of 50 ml, with 13 TaKaRa Taq Buffer with Mg21, 200 lM of each dNTP, 200 nM of forward and reverse primers, 1.25 U of TaKaRa Taq, and 50–100 ng of DNA template). PCR cycling parameters were as follows: an initial denaturation at 958C for 5 min followed by 35 cycles at 958C for 30 seconds, 55–618C for 45 seconds, 728C for 30 seconds; and a final extension at 728C for 10 min. Five microliters of PCR product was run on a 1.5% agarose gel to check the specificity of amplification, and 20 ml was sent for direct sequencing with forward and reverse primers for each gene (SinoGenoMax Co., Ltd., China).
Statistical Analysis The histological diagnosis in each patient was obtained from their medical records. The results of histopathology were taken as the gold standard. Cytological and molecular results were compared with histopathology. Group comparisons of categorical variables were performed using the v2 test. The association between clinicopathologic factors and presence of the BRAF p.V600E mutation was validated by the Pearson v2 and independent t tests. Statistical analysis was carried out using SPSS 13.0 and P < 0.05 was considered statistically significant.
Results A total of 195 FNA samples from 44 male and 151 female patients (mean age: 45.1 years), were included in this study. All FNA specimens were sufficient to successfully perform DNA extraction and PCR analysis. On surgical follow-up, 170 cases were confirmed as PTC, and the lesions in the other 25 cases were benign. Using cytology alone, the diagnostic sensitivity, specificity, and accuracy for the detection of PTC were 70% (119/170), 100% (25/25), and 73.8% (144/195), respectively. BRAF p.V600E testing showed similar sensitivity (67.1%), specificity (100%), and accuracy (71.3%) to that Diagnostic Cytopathology, Vol. 43, No 10
787
Diagnostic Cytopathology DOI 10.1002/dc
ZHAO ET AL. Table I. Diagnostic Values of FNA Cytology, BRAF p.V600E Testing or a Combination of Both for Detecting PTC Diagnostic modality Cytologya Cytologyb BRAF p.V600E testing Combineda Combinedb
SN (%)
SP (%)
PPV (%)
NPV (%)
AC (%)
70.0 91.8 67.1
100.0 84.0 100.0
100.0 97.5 100.0
32.9 60.0 30.9
73.8 90.8 71.3
85.3 92.9
100.0 84.0
100.0 97.5
50.0 63.6
87.2 91.8
a
Cytology diagnosis of malignant was taken as the diagnostic threshold. b Cytology diagnosis of suspicious malignant was taken as the diagnostic threshold. Abbreviations: SN, sensitivity; SP, specificity; PPV, positive predictive value; NPV, negative predictive value; AC, accuracy. Table II. Correlation Between Cytologic Diagnoses and BRAF p.V600E Mutation Status
Cytologic diagnoses (n) PTC (n 5 170)
Total Benign (n 5 25)
Total
PTC SPTC FN AUS Benign SPTC FN AUS Benign
119 37 2 7 5 170 4 2 1 18 25
BRAF p.V600E (1) (n)
(%)
88 24 0 2 0 114 0 0 0 0 0
73.9 64.9 0 28.6 0 67.1 0 0 0 0 0
Abbreviations: AUS, atypia of undetermined significance; FN, follicular neoplasm; SPTC, suspicious papillary thyroid carcinoma; PTC: papillary thyroid carcinoma.
of cytology. When FNA cytology was combined with BRAF p.V600E testing, improved sensitivity (85.3% vs 70.0%, P 5 0.001) and accuracy (87.2% vs 73.8%, P 5 0.001) were obtained for PTC diagnosis (Table I). Table II shows the correlation between BRAF p.V600E mutations and cytological diagnoses. In terms of the two cytology categories of SPTC and AUS, BRAF p.V600E testing contributed to cytology diagnosis. In the group with SPTC (n 5 41), 24 samples had BRAF p.V600E mutations. At final histology, 37 were confirmed as PTC by histopathology and all BRAF p.V600E mutated cases were histologically confirmed as PTC. BRAF p.V600E testing confirmed 64.9% (24/37) of PTC with SPTC cytology diagnosis. In the group with AUS cytology (n 5 8), 7 were PTC. Two samples had BRAF p.V600E mutations, and both were confirmed as PTC. BRAF p.V600E testing detected 28.6% (2/7) of PTC in AUS cytology samples. The clinical pathologic features of PTC according to BRAF p.V600E mutation status are shown in Table III. Of 170 PTCs, 13 recurrent patients (patients who already had a diagnosis of papillary carcinoma) were excluded from this 788
Diagnostic Cytopathology, Vol. 43, No 10
Table III. Univariate Analysis of the Association Between BRAF p.V600E Mutation and Clinicopathologic Factors BRAF p.V600E Positive (n 5 105) Characteristic
n
%
Age at diagnosis (years) Mean 6 SD 45.60 6 10.89 Range 21–76 Gender Male 23 21.9 Female 82 Capsular invasion Present 83 79 Absent 22 Extra-thyroid extension Present 49 46.7 Absent 56 Lymph node metastasis Present 54 51.5 Absent 51 Tumor size 1 cm 34 32.4 >1 cm 71 Multifocality Yes 41 39.0 No 64 AJCC stages I/II 78 74.3 III/IV 27
BRAF p.V600E Negative (n 5 52) n
%
39.8 6 10.80 18–67
P 0.002*
11 41
21.2
0.914
41 11
78.8
0.977
22 30
42.3
0.606
30 22
57.7
0.459
16 36
30.8
0.838
18 34
34.6
0.589
42 10
80.8
0.368
study. Of the other 157 cases, 123 were female and 34 were male aged 18–76 years, with an average age of 44 years. Patients with BRAF p.V600E positive tumors were significantly older than those who did not harbor the mutation (45.6 years vs. 39.8 years, P 5 0.002). However, BRAF p.V600E mutation was not associated with patient gender, capsular invasion, extra-thyroid extension, lymph node metastasis, tumor size, multifocality, and TNM stage. In addition, no significant relationships between BRAF p.V600E mutation and clinicopathologic characteristics were observed in the PTC group with tumor size 1 cm (Table IV).
Discussion It has been reported that BRAF p.V600E gene mutations can be successfully identified using liquid-based material with high reproducibility.20 Therefore, it is necessary to establish criteria regarding adequate cell number for molecular testing and to avoid false-negative findings. Previous studies suggested selecting 6–10 groups of cells on the slide to perform PCR detection.19 In order to guarantee enough DNA for analysis, we chose 10 groups of cells as the criterion in this study. With regard to the correct volume of tumor sample, most studies do not mention this. In this study, we were able to extract 300 ng to [mt]110,000 ng DNA from 10 ml of residual routine FNA specimens. This was sufficient to perform PCR detection. As this was an initial study on liquid-based residual specimens, further studies should be performed to
Diagnostic Cytopathology DOI 10.1002/dc
BRAF P.V600E TESTING IN THYROID FNA Table IV. Univariate Analysis of the Association Between BRAF p.V600E Mutation and Clinicopathologic Factors of PTCs with Tumor Size 1 cm BRAF p.V600E Positive (n 5 34) Characteristic
n
%
Age at diagnosis (years) Mean 6 SD 47.44 6 10.56 Range 25–65 Gender Male 6 17.6 Female 28 Capsular invasion Present 15 44.1 Absent 19 Extra-thyroid extension Present 11 32.4 Absent 23 Lymph node metastasis Present 12 35.3 Absent 22 Multifocality Yes 8 23.5 No 26 AJCC stages I/II 26 76.5 III/IV 8
BRAF p.V600E Negative (n 5 16) n
%
41.69 6 9.36 28–56
P 0.069
3 13
18.8
0.925
7 9
43.8
0.981
3 13
18.8
0.318
5 11
31.3
0.778
5 11
31.3
0.562
14 2
87.5
0.363
determine the smallest residual sample which can be used for molecular detection. BRAF p.V600E mutation is known to be highly specific to PTC and has not been reported in follicular carcinoma, adenoma, or benign thyroid nodules.21 Due to its high specificity, this genetic alteration is now considered a useful diagnostic marker for thyroid FNA. Zatelli et al. reported that BRAF p.V600E mutation analysis increased the sensitivity of cytology for PTC from 77% to 87%.22 Another author observed that the rate of PTC diagnosis using cytology alone was 62.3% (56/90), while the rate of PTC diagnosis using a morphomolecular method was 82.2% (74/90), with an increase of 20% in diagnostic accuracy due to BRAF p.V600E mutation analysis.23 In our study, we found that BRAF p.V600E mutation testing increased the sensitivity of the FNA procedure from 70% to 85.3%. Therefore, the addition of molecular analysis resulted in an increase in sensitivity of 15.3% compared to cytology alone. Diagnostic accuracy also increased from 73.8% to 87.2%. These findings are further evidence of the efficacy of BRAF p.V600E as a diagnostic marker in thyroid FNA. In the study by Xing, the authors believed that if BRAF p.V600E testing was performed before cytology examination, nearly half of the patients with PTC did not need to undergo cytology examination, which could result in a substantial cost saving.24 However, in clinical practice, the majority (60–80%) of FNA specimens are benign lesions and the risk of malignancy in cytology benign lesions is estimated to be
