BRAF V600E detection in cytological thyroid samples: A key component of the decision tree for surgical treatment of papillary thyroid carcinoma

Transcription

1 ORIGINAL ARTICLE BRAF V600E detection in cytological thyroid samples: A key component of the decision tree for surgical treatment of papillary thyroid carcinoma Jean-Franc ois Collet, MD, 1,2 Roger Lacave, MD, PhD, 3 Sylvain Hugonin, AD, 3 Virginie Poulot, AD, 3 Marc Tassart, MD, 4 Anne Fajac, MD, PhD 1 * 1 Anatomie et Cytologie Pathologiques, H^opital Tenon HUEP, APHP, Paris, France, 2 Institut de Pathologie, Paris, France, 3 Unite degenomique des Tumeurs Solides, H^opital Tenon HUEP, APHP, Paris, France, 4 Radiologie, H^opital Tenon, HUEP, APHP, Paris, France. Accepted 17 December 2015 Published online 8 February 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI /hed ABSTRACT: Background. Whether preoperative knowledge of the BRAF mutation status would help to determine the extent of surgery for thyroid nodules is still under investigation. Methods. We developed a method to state the V600E mutation before surgery on fine-needle aspiration (FNA) stained smears checked to contain tumor cells. We evaluated the interest of the preoperative assessment of the mutation for surgical strategy of nodules, diagnosed as malignant, suspicious for malignancy or follicular neoplasms. Results. The mutation was found in 81% (79 of 97) malignant, 59% (20 of 34) suspicious nodules, and in none of follicular neoplasms (n 5 29). Overall, the mutation was detected in 82% of papillary carcinomas. The sensitivity, specificity, and positive and negative predictive values for the diagnosis of malignancy were 75%, 100%, 100%, and 46%, respectively. Conclusion. The preoperative knowledge of the V600E mutation status is fundamental to plan total thyroidectomy with certainty and should be part of the decision tree for the management of thyroid nodules. VC 2016 Wiley Periodicals, Inc. Head Neck 38: , 2016 KEY WORDS: BRAF, papillary thyroid cancer, fine-needle aspiration (FNA), cytology, preoperative diagnosis INTRODUCTION Papillary thyroid carcinoma is the most common histological type of thyroid cancer. The overall prognosis of this cancer is good with a 10-year survival rate >90%. However, the recurrence rate, especially in neck lymph nodes, may be up to 20% to 30%. 1,2 The optimal surgical treatment of papillary thyroid carcinoma remains controversial. Although total thyroidectomy is well-established, there is no agreement on the extent of lymph node dissection whether therapeutic or prophylactic. 3,4 Reoperation per se of patients with papillary thyroid carcinoma has been recently shown to be an independent factor of poor prognosis. 5 Therefore, there is a need for factors to guide the extent of initial surgery. There are already well-known clinicopathological prognosis factors in papillary thyroid carcinoma, including age, sex, tumor size, and stage of the tumor. 6 However, risk stratification based upon these factors is incomplete and often inapplicable before surgery. So far, the molecular event most often reported in papillary thyroid carcinoma is a mutation in the proto-oncogene BRAF, specifically the V600E mutation, with an average rate of 49% in a meta-analysis. 7 This mutation has been reported to be a poor prognosis factor in several studies, including *Corresponding author: A. Fajac, Anatomie et Cytologie Pathologiques, H^opital Tenon HUEP, APHP, 4 rue de la Chine, Paris, France. anne.fajac@.aphp.fr a study with a long follow-up In addition, it has been suggested that the preoperative knowledge of the BRAF mutation status may help to guide the extent of initial surgery. 11,12 However, whether this mutation is a useful molecular marker and should be routinely tested for are not precisely stated. 13,14 The goals of the present study were (1) to develop a reliable method on fine-needle aspiration (FNA) samples in order to provide the V600E mutation status before surgery and (2) to evaluate separately the interest of the V600E mutation in thyroid nodules diagnosed as 1/malignant, 2/suspicious for malignancy, or 3/follicular neoplasms by FNA (ie, categories VI, V, and IV according to the National Cancer Institute/Bethesda classification). 15 To avoid false-negative results because of the absence of tumor cells in the analyzed sample, a stained smear checked to contain tumor cells was used for the molecular analysis. MATERIALS AND METHODS Thyroid fine-needle aspiration samples Between 2011 and 2014, a consecutive series of thyroid nodules that were diagnosed by FNA as papillary carcinoma (n 5 93) were preoperatively tested for the presence of the V600E mutation. Nodules diagnosed as other malignancy (n 5 4), as suspicious for malignancy (n 5 34), and as follicular neoplasms, including the oxyphilic neoplasms (n 5 29), respectively categories VI, V, and IV, according to the National Cancer Institute/Bethesda HEAD & NECK DOI /HED JULY

2 COLLET ET AL. TABLE 1. Patients characteristics (n 5 160). Characteristics patients Sex: F/M 124/36 Age mean, y (range) 48 (22 81) TN classification* T1 Total n 5 46 T1N0 17 T1N1a 14 T1N1b 8 T1Nx 7 T2 Total n 5 6 T2N0 3 T2N1a 1 T2N1b 1 T2Nx 1 T3 Total n 5 33 T3N0 12 T3N1a 15 T3N1b 4 T3Nx 2 T4 Total n 5 4 T4N1a 1 T4N1b 2 T4Nx 1 * TN classification was available for 89 patients diagnosed with papillary carcinoma. classification 15 were also tested. The Bethesda IV category included 21 follicular neoplasms and 8 oxyphilic neoplasms. Histological diagnosis was available for all the cases. Patients characteristics are specified in Table 1. No lymph node FNA sample was included in the present series. Preparation of the fine-needle aspiration samples for polymerase chain reaction FNA was performed either directly or ultrasoundguided, using very fine needles (25G/27G), without aspiration. No local anesthesia and interruption or changes of anticoagulant treatment if any were required. The material obtained from FNA was directly spread on slides. Smears were air-dried and stained according to the May Grunwald Giemsa method for cytological diagnosis. To test for the V600E BRAF mutation, a stained smear checked to contain tumor cells was destained with 70% methanol for at least 2 hours and then gently scraped (see Figure 1). The number of tumor cells present in the smears analyzed for the mutation was semiquantitatively evaluated as low (1), moderate (11), or high (111). Most of the smears were 11 or 111. Nontumor cells, such as macrophages or blood cells, were present in the smears but in low proportions, as the mean percentage of tumor cells was 90% (range, 50% to 95%). One smear per patient was sufficient for V600E testing. Polymerase chain reaction DNA was extracted using the Qiagen kit, in accord with the manufacturer s instructions. DNA (25 ng) was amplified using 2 specific primers and 2 TaqMan-specific probes, whose sequences are mentioned below, for allelic discrimination of the V600E mutation (T1799A): BRAF-F: 5 0 -CTACTGTTTTCCTTTACTTACTACACC TCAGA BRAF-R: 5 0 -ATCCAGACAACTGTTCAAACTGATG Wild-type probe: 5 0 -VIC-CTAGCTACAGTGAAATC-3 0 Mutant probe: 5 0 -FAM-TAGCTACAGAGAAATC-3 0. Characterized V600E mutated human colorectal adenocarcinoma HT29 cell line (ATCC number HTB-38) was used as standard positive control with a 5% sensitivity threshold when diluted in nonmutated cells (human mononuclear cells). External quality assessment was certified by annual participation to the French EQA BRAF melanoma ( 16 RESULTS Feasibility to detect the V600E BRAF mutation in stained smears The possibility to detect the V600E BRAF mutation was evaluated in stained smears morphologically checked to contain papillary carcinoma cells. In these preliminary experiments, detection of the mutation was demonstrated in 7 cases. For 4 of these positive cases, the V600E mutation was also tested in the thyroid tumor obtained after surgery from slides containing unstained paraffinembedded tissue to control that the detection of the mutation in FNA stained smears was not an artifact because of the staining. The mutation was present in all the histological cases. The sensitivity of the polymerase chain reaction (PCR) amplification for detection of the V600E BRAF mutation was 5% tumor cells (see Materials and Methods). BRAF V600E mutation in fine-needle aspiration samples preoperatively diagnosed as malignant, suspicious for malignancy, or follicular neoplasms The results are summarized in Table 2. A preoperative diagnosis of malignancy (Bethesda VI category) was made for 97 nodules, including a majority of papillary FIGURE 1. Procedure for BRAF V600E testing from thyroid fine-needle aspiration (FNA) smears. PCR, polymerase chain reaction. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com] 1018 HEAD & NECK DOI /HED JULY 2016

3 BRAF MUTATION DETECTION IN THYROID CYTOLOGICAL SAMPLES TABLE 2. BRAF V600E mutation status according to Bethesda category and cytological diagnosis (n 5 160). Bethesda category nodules nodules with V600E (%) Cytological diagnosis nodules nodules with V600E VI (malignant) (81) Papillary carcinoma Follicular carcinoma 2 0 Medullary carcinoma 1 0 V (suspicious) (59) For papillary carcinoma For malignancy NOS 11 5 IV (follicular neoplasms) 29 0 (0) Follicular neoplasms 21 0 Oxyphilic neoplasms 8 0 Abbreviation: NOS, not otherwise specified. carcinomas (n 5 93). The V600E mutation rate for malignant cases was 81% (79 of 97). A diagnosis of suspicious for malignancy (Bethesda V category) was established for 34 nodules. It could be more precisely defined as suspicious for papillary carcinoma for 23 cases. The mutation rate for suspicious nodules was 59% (20 of 34). No V600E mutation was detected in the 29 nodules preoperatively diagnosed as follicular neoplasms, including the oxyphilic neoplasms (Bethesda IV category). Evaluation of the BRAF V600E mutation as a biomarker of malignancy The results of the V600E mutation status, according to the final histological diagnosis established on formalinfixed paraffin-embedded tissue specimens after surgical resection, are summarized in Table 3. A total of 120 cases of papillary carcinoma were diagnosed by histology. The cytological diagnosis of malignancy was confirmed for all the 97 cases. When the V600E mutation was detected, it was evidenced exclusively in papillary carcinoma (n 5 79). Concerning the suspicious nodules, the final histological diagnosis was malignancy in most cases (n 5 28; 82%). Again, in all the suspicious nodules for which the V600E mutation was detected, the final histopathological diagnosis was papillary carcinoma (n 5 20). The follicular neoplasms, for all of which no mutation was detected, were diagnosed as benign by histology in most cases (n 5 22; 76%). Malignant cases were follicular or oxyphilic carcinoma with no papillary carcinoma. Overall, no mutation was detected in benign tumors or in other malignancies than papillary carcinoma. The V600E mutation rate was 82% (99 of 120) for papillary carcinomas, including 88 (of 100) conventional papillary carcinomas, 6 (of 12) follicular variants, and 5 (of 6) TABLE 3. BRAF V600E mutation status according to histological diagnosis (n 5 160). Bethesda category nodules Cytological diagnosis nodules Histological diagnosis nodules* nodules with V600E VI (malignant) 97 Papillary carcinoma 93 Papillary carcinoma 91* 79 Medullary carcinoma 1 0 Follicular carcinoma 2 Follicular carcinoma 1 0 Papillary carcinoma 1* 0 Medullary carcinoma 1 Medullary carcinoma 1 0 Undifferentiated carcinoma 1 Papillary carcinoma 1* 0 V (suspicious) 34 For papillary carcinoma 23 Papillary carcinoma 19* 15 Oxyphilic adenoma 2 0 Trabecular adenoma 1 0 For malignancy NOS 11 Papillary carcinoma 8* 5 Oxyphilic adenoma 1 0 Follicular adenoma 1 0 Lymphocytic thyroiditis 1 0 IV (follicular neoplasms) 29 Follicular neoplasms 21 Follicular carcinoma 2 0 Oxyphilic carcinoma 1 0 Follicular adenoma 15 0 Oxyphilic adenoma 3 0 Oxyphilic variant 8 Oxyphilic carcinoma 4 0 Oxyphilic adenoma 4 0 Abbreviation: NOS, not otherwise specified. * Total number of papillary carcinoma n Total number of nodules with the BRAF V600E mutation n HEAD & NECK DOI /HED JULY

4 COLLET ET AL. oxyphilic variants. No mutation was detected in the 2 tall-cell variants of the present series. The sensitivity, specificity, and positive and negative predictive values of V600E mutation for the diagnosis of malignancy were 75%, 100%, 100%, and 46%, respectively. DISCUSSION In the present study, we show that (1) the preoperative determination of BRAF V600E mutation status of thyroid nodules is possible directly from the stained FNA smears checked to contain tumor cells, that (2) the percentage of papillary carcinoma with V600E mutation is high (around 80%), and that (3) the positive predictive value of this mutation for the diagnosis of papillary thyroid carcinoma is 100%. The procedure for the testing of the V600E mutation reported here avoids false-negative results because of the absence of tumor cells in the analyzed sample. It is easy and quick as results can be obtained within 2 days after the cytological diagnosis, therefore within the time frame for scheduling thyroid surgery. As opposed to immunohistochemical techniques, which have been developed for the detection of the V600E mutation, the present technique is not subjected to the observer interpretation. Equivocal staining has indeed been reported for V600E immunodetection as well as false-positive and falsenegative results In the present series, the V600E mutation rate in papillary thyroid carcinoma is higher than the mean rate commonly reported; around 50%. 7 The reasons for this high prevalence are unclear. It may be due to the method we used, which decreases the percentage of false-negative results because of the presence of tumor cells in all the samples tested for the V600E mutation, to the high proportion of tumor cells versus nontumor cells (mean, 90%; see Materials and Methods), and to the high sensitivity of the PCR technique (level of detection: 5% tumor cells). Given the high V600E mutation rate, we doubt that presence of the mutation could constitute a marker of poor prognosis and could guide the extent of lymph node dissection. However, the presence of the mutation indicates that a total thyroidectomy can be performed with certainty avoiding unnecessary examination, which is systematically performed in some centers even if a preoperative cytological diagnosis of papillary carcinoma has been made. On the opposite side, whether the absence of the mutation is an indicator of good prognosis with a low recurrence rate remains to be determined. However, potentially more interesting is the fact that, among the suspicious nodules (Bethesda V category), all nodules with the V600E mutation were papillary carcinoma at final histological diagnosis. Utility of BRAF testing for the subgroup of suspicious nodules has been previously reported but with no such high specificity and positive predictive value for the diagnosis of papillary carcinoma as in the present series. 20 We propose that the preoperative diagnosis of V600E mutation in a suspicious nodule should allow total thyroidectomy at once. This would avoid the need for unnecessary examination or for a 2-step surgery in case of a diagnostic hemithyroidectomy with a final diagnosis of malignancy. In our institution, the cost of the V600E BRAF test is $65 U.S. dollars. The routine screening of V600E mutation for suspicious nodules might be cost effective, as recently suggested by a Dutch group. 21 No V600E mutation was found in nodules diagnosed as follicular or oxyphilic neoplasms by FNA. For the cases that were malignant at final histological diagnosis, including follicular and oxyphilic carcinoma, the absence of V600E mutation is in accordance with the fact that these types of thyroid carcinoma exhibit other molecular alterations during carcinogenesis, such as ras mutation. 22 In conclusion, a high prevalence of V600E mutation was found in papillary thyroid carcinoma in the present series using a preoperative procedure optimized to avoid false-negative results. We think that the preoperative knowledge of V600E mutation status is fundamental to plan with certainty a total thyroidectomy in case the V600E mutation is present, whether the cytological diagnosis is papillary carcinoma or mainly in case of suspicious for malignancy. We therefore believe that assessment of the BRAF mutation status should be part of the decision tree for the management of thyroid nodules. REFERENCES 1. Simon D, Goretzki PE, Witte J, R oher HD. Incidence of regional recurrence guiding radicality in differentiated thyroid carcinoma. World J Surg 1996;20: ; discussion Hay ID, McConahey WM, Goellner JR. Managing patients with papillary thyroid carcinoma: insights gained from the Mayo Clinic s experience of treating 2,512 consecutive patients during 1940 through Trans Am Clin Climatol Assoc 2002;113: McLeod DS, Sawka AM, Cooper DS. Controversies in primary treatment of low-risk papillary thyroid cancer. Lancet 2013;381: Lang BH, Ng SH, Lau LL, Cowling BJ, Wong KP, Wan KY. A systematic review and meta-analysis of prophylactic central neck dissection on shortterm locoregional recurrence in papillary thyroid carcinoma after total thyroidectomy. Thyroid 2013;23: Young S, Harari A, Smooke Praw S, Ituarte PH, Yeh MW. Effect of reoperation on outcomes in papillary thyroid cancer. Surgery 2013;154: ; discussion Tubiana M, Schlumberger M, Rougier P, et al. Long-term results and prognostic factors in patients with differentiated thyroid carcinoma. Cancer 1985;55: Lee JH, Lee ES, Kim YS. Clinicopathologic significance of BRAF V600E mutation in papillary carcinomas of the thyroid: a meta-analysis. Cancer 2007;110: Xing M, Westra WH, Tufano RP, et al. BRAF mutation predicts a poorer clinical prognosis for papillary thyroid cancer. J Clin Endocrinol Metab 2005;90: Elisei R, Ugolini C, Viola D, et al. BRAF(V600E) mutation and outcome of patients with papillary thyroid carcinoma: a 15-year median follow-up study. J Clin Endocrinol Metab 2008;93: Prescott JD, Sadow PM, Hodin RA, et al. BRAF V600E status adds incremental value to current risk classification systems in predicting papillary thyroid carcinoma recurrence. Surgery 2012;152: Yip L, Nikiforova MN, Carty SE, et al. Optimizing surgical treatment of papillary thyroid carcinoma associated with BRAF mutation. Surgery 2009;146: Xing M, Clark D, Guan H, et al. BRAF mutation testing of thyroid fineneedle aspiration biopsy specimens for preoperative risk stratification in papillary thyroid cancer. J Clin Oncol 2009;27: Handkiewicz Junak D, Czarniecka A, Jarzab B. Molecular prognostic markers in papillary and follicular thyroid cancer: current status and future directions. Mol Cell Endocrinol 2010;322: Lee KC, Li C, Schneider EB, et al. Is BRAF mutation associated with lymph node metastasis in patients with papillary thyroid cancer? Surgery 2012;152: Cibas ES, Ali SZ; NCI Thyroid FNA State of the Science Conference. The Bethesda system for reporting thyroid cytopathology. Am J Clin Pathol 2009;132: Emile JF, Tisserand J, Bergougnoux L, et al. Improvement of the quality of BRAF testing in melanomas with nationwide external quality assessment, for the BRAF EQA group. BMC Cancer 2013;13: Zimmermann AK, Camenisch U, Rechsteiner MP, Bode Lesniewska B, R ossle M. Value of immunohistochemistry in the detection of BRAF(V600E) mutations in fine-needle aspiration biopsies of papillary thyroid carcinoma. Cancer Cytopathol 2014;122: HEAD & NECK DOI /HED JULY 2016

5 BRAF MUTATION DETECTION IN THYROID CYTOLOGICAL SAMPLES 18. Rossi ED, Martini M, Capodimonti S, et al. Analysis of immunocytochemical and molecular BRAF expression in thyroid carcinomas: a cytohistologic institutional experience. Cancer Cytopathol 2014;122: Lee SR, Yim H, Han JH, et al. VE1 antibody is not highly specific for the BRAF V600E mutation in thyroid cytology categories with the exception of malignant cases. Am J Clin Pathol 2015;143: Jara SM, Bhatnagar R, Guan H, Gocke CD, Ali SZ, Tufano RP. Utility of BRAF mutation detection in fine-needle aspiration biopsy samples read as suspicious for papillary thyroid carcinoma. Head Neck [Epub ahead of print] 21. Lodewijk L, Kist JW, Valk GD, Vriens MR, Borel Rinkes IH. Comment on preoperative BRAF(V600E) mutation screening is unlikely to alter initial surgical treatment of patients with indeterminate thyroid nodules: a prospective case series of 960 patients. Cancer 2014;120: Xing M. Molecular pathogenesis and mechanisms of thyroid cancer. Nat Rev Cancer 2013;13: HEAD & NECK DOI /HED JULY