Molecular testing of BRAF, RAS and TERT on thyroid FNAs with indeterminate cytology improves diagnostic accuracy

Transcription

1 Accepted: 27 September 2017 DOI: /cyt ORIGINAL ARTICLE Molecular testing of BRAF, RAS and TERT on thyroid FNAs with indeterminate cytology improves diagnostic accuracy M. Decaussin-Petrucci 1,6 * F. Descotes 2 * L. Depaepe 1 V. Lapras 3 M.-L. Denier 3 F. Borson-Chazot 4 J.-C. Lifante 5 J. Lopez 2,6 1 Pathology department, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon, Pierre Benite, France 2 Biochemistry and molecular biology department, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon, Pierre Benite, France 3 Radiology department, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon, Pierre Benite, France 4 Endocrinology department, Groupement hospitalier Est, Hospices Civils de Lyon, Bron, France 5 Endocrine surgery department, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon, Pierre Benite, France 6 Cancer Research Center of Lyon, INSERM1052 CNRS5286, Universite de Lyon, Lyon, France Correspondence Jonathan Lopez, Service de Biochimie et Biologie moleculaire, Centre Hospitalier Lyon Sud, Pierre-Benite, France. jonathan.lopez@univ-lyon1.fr Objective: Liquid-based (LB)-FNA is widely recognized as a reliable diagnostic method to evaluate thyroid nodules. However, up to 30% of LB-FNA remain indeterminate according to the Bethesda system. Use of molecular biomarkers has been recommended to improve its pathological accuracy but implementation of these tests in clinical practice may be difficult. Here, we evaluated feasibility and performance of molecular profiling in routine practice by testing LB-FNA for BRAF, N/ HRAS and TERT mutations. Methods: We studied a large prospective cohort of 326 cases, including 61 atypia of undetermined significance, 124 follicular neoplasms, 72 suspicious for malignancy and 69 malignant cases. Diagnosis of malignancy was confirmed by histology on paired surgical specimen. Results: Mutated LB-FNAs were significantly associated with malignancy regardless of the cytological classification. Overall sensitivity was 60% and specificity 89%. Importantly, in atypia of undetermined significance and follicular neoplasm patients undergoing surgery according to the Bethesda guidelines, negative predictive values were 85.4% and 90% respectively. TERT promoter mutation was rare but very specific for malignancy (5.5%) suggesting that it could be of interest in patients with indeterminate cytology. Conclusions: Mutation profiling can be successfully performed on thyroid LB-FNA without any dedicated sample in a pathology laboratory. It is an easy way to improve diagnostic accuracy of routine LB-FNA and may help to better select patients for surgery and to avoid unnecessary thyroidectomies. KEYWORDS indeterminate cytology, molecular testing, TERT promoter mutations, thyroid cancer, thyroid liquid-based fine needle aspiration 1 INTRODUCTION Thyroid nodules are very common, with a 5%-7% clinical prevalence among the European population, markedly increasing in iodinedeficient countries. 1,2 The majority of thyroid nodules are benign, with thyroid cancer accounting for less than 7% of nodules. 3 *These two authors contributed equally to the work Liquid-based (LB)-FNA is the most accurate and cost-effective method to routinely evaluate of thyroid nodules before surgery. LB- FNA allows a definitive benign or malignant diagnosis in 60%-80% of cases. 3,4 However, up to 30% of adequate LB-FNA are indeterminate, and patients are referred to surgery for diagnostic thyroidectomy. 5 According to the Bethesda system, the indeterminate category includes atypia of undetermined significance/follicular lesion of undetermined significance (AUS/FLUS), follicular neoplasm/suspicious for follicular Cytopathology. 2017;1 6. wileyonlinelibrary.com/journal/cyt 2017 John Wiley & Sons Ltd 1

2 2 DECAUSSIN-PETRUCCI ET AL. neoplasm or H urthle cell neoplasm/suspicious for H urthle cell neoplasm (FN/SFN), and suspicious for malignancy (SM). 6 Recently, molecular testing has been proposed to refine the assessment of cancer risk in nodules with indeterminate cytology. The most studied alterations are BRAF and RAS mutations, and RET/ PTC and PAX8/PPARg rearrangements. 7 BRAF mutation is highly specific of papillary thyroid cancer (PTC), present in 29%-77% of PTC, but the sensitivity is too low to reliably rule out the presence of malignancy. 8 PAX8/PPARg translocation is present in follicular carcinoma with frequencies up to 30%, but also to a lesser extent in some follicular variant of PTC and rarely in follicular adenoma. RET/ PTC rearrangements are found in 10%-20% of PTC with a higher frequency in children/young adults and in patients with a history of radiation exposure. RAS-family mutations are less specific of thyroid carcinoma, with a lower positive predictive value than BRAF, as these alterations are also detected with a variable percentage of non-malignant lesions. 9 More recently, a gene expression classifier (GEC) test (AFIRMA) and a next-generation sequencing panel (Thyroseq v2.1) were also proposed. 10,11 These tests are very expensive, and currently only available in the USA. Furthermore, most of the time all these molecular tests requires an additional frozen LB- FNA dedicated to the molecular study, difficult to obtain in daily practice, especially as the diagnosis of indeterminate category is not known at the moment of the LB-FNA. Therefore, most of the time they are the reserve of university hospital centres, although the American Thyroid Association (ATA) guidelines recommend incorporating molecular testing in the algorithms used to manage patients with indeterminate thyroid nodules to limit unnecessary surgery. 12 Recently, TERT promoter mutations were proposed as a prognostic molecular marker in thyroid carcinomas. They were predominantly found in more aggressive diseases, such as tall cell variant of PTC, widely invasive follicular carcinoma, poorly differentiated carcinoma and anaplastic carcinoma. By contrast, TERT mutation was absent in 363 benign thyroid nodules except one, 13 demonstrating a high specificity for malignant neoplasms. Prevalence of TERT promoter mutations is variable across countries, with rates ranging from 7.5% to 25.5% (median 11.9%) for PTC and from 13.8% to 36.4% for follicular thyroid carcinoma (FTC) Performance of TERT mutations as a diagnostic marker in LB- FNA was never described in the European population, where the incidence of mutations is different from Asia and North America. 15,19,20 Here, we evaluated feasibility and performance of molecular profiling in routine practice by testing residual material from LB-FNA for BRAF, NRAS, HRAS and TERT mutations in a large prospective cohort of 326 cases with indeterminate or malignant cytology, using histology on paired surgical specimens as gold standard to assess diagnosis of malignancy. Civils de Lyon, Lyon, France) in a prospective study. LB-FNAs were classified as indeterminate (AUS/FN/SM) and malignant (M) according to the 2010 Bethesda system. Two pathologists (M.D.P. and L.D.) independently reviewed all the cases. All patients underwent thyroid surgery according to the Bethesda guidelines. To limit bias, molecular testing results were not communicated to clinicians prior to surgery. Haematoxylin-phloxine-saffron-stained slides of the surgical specimens were reviewed in all cases (M.D.P.), to confirm diagnosis and classify lesions according to the 2004 WHO classification (DeLellis RA et al 2004). All the follicular variants of papillary carcinomas were reviewed to identify non-invasive neoplasms with papillary-like features (NIFTP) according to the refine diagnostic criteria established by Nikiforov and al. 21 In this study we classified this neoplastic variant in the malignant group. Histopathological diagnoses were blinded to the molecular testing results to limit bias. The ethics committee of the medical faculty and the state medical board all agreed to these investigations and an informed consent was obtained for all patients included in this study. 2.2 Molecular analysis To reduce cost, facilitate integration in routine clinical practice and avoid an additional FNA for molecular testing, we used residual material of thyroid LB-FNA after cytology examination and a very simple four-gene panel. When quantity of DNA was insufficient to perform the whole panel, analyses were prioritised as follow: N/ HRAS in AUS and FN and BRAF in SM and M categories. DNA was extracted using Circulating Nucleic Acid Kit QIAamp â according to the supplier s recommendations (Qiagen, Hilden, Germany). Mutations for BRAF, NRAS and HRAS were screened by high resolution melt approach on a LightCycler LC480 instrument (Roche Diagnostics, Mannheim, Germany) and confirmed by Sanger sequencing. Mutations of TERT promoter were analysed by nested polymerase chain reaction (PCR) and Sanger sequencing. The first PCR (forward 5 0 -CACCCGTCCTGCCCCTTCACCTT-3 0 and reverse 5 0 -GGCTTCCCA CGTGCGCAGCAGGA-3 0 ) amplifies a 275-bp fragment used as matrix for the second PCR (forward 5 0 -CCCCTTCACCTTCCAGCTC-3 0 and reverse 5 0 -GCCGCGGAAAGGAAGG-3 0 ) amplifying a fragment of 118-bp carrying the point mutations -124 (C228T) and -146 (C250T). PCR products were then sequenced according to the Sanger method. Internal negative (ie normal thyroid tissue and thyroid benign lesions) and positive (ie mutated) controls were included in all our molecular testing series. 3 RESULTS 2 MATERIALS AND METHODS 2.1 Patients and samples The study included 326 specimens of liquid-based fine needle aspiration (LB-FNA) obtained from Lyon Sud University Hospital (Hospices According to the Bethesda System the 326 cases were classified as 61 AUS/FLUS, 124 FN/SFN, 72 SM and 69 M. Among the 326 cases, 163 were found to be malignant by histology on paired surgical specimen. Mutation profiling of BRAF, NRAS, HRAS and TERT was performed prospectively before surgery on residual material from

3 DECAUSSIN-PETRUCCI ET AL. 3 thyroid LB-FNA in this large cohort. Overall mutation rate was 35.6% (116 cases). Sensitivity of this routine molecular panel to detect malignancy was 60.1% (98/163) and specificity 89.0% (145/ 163). BRAF was mutated in 44.8% (n = 73) of the 163 LB-FNAs confirmed to be malignant, NRAS/HRAS in 14.7% (n = 24) and TERT promoter in 5.5% (n = 9). Importantly, BRAF/RAS/TERT profiling was significantly associated with malignancy independently of the Bethesda category. Repartition of mutations and comparison between benign and malignant histology in each Bethesda category is shown in Figure 1 and Table S1. Focusing on the AUS and FN groups, NRAS/HRAS mutations were the most frequent, found in both malignant and benign neoplasms but significantly more frequently in the malignant ones only in the AUS group (AUS: 64.7% vs 18.2%, P =.0013; FN: 18.8% vs 8.3%, P =.1865, ns). Of note, positive predictive values were, respectively, 57.9% (11/19) for AUS and 25% (3/12) for the FN category. TERT mutations were found in three LB-FNAs classified as FN and confirmed to be malignant. Interestingly, two of them were not associated with any BRAF or RAS mutation. All the mutated samples from the SM group were confirmed to be malignant, emphasizing a high specificity of our fourgene panel in this category. Overall sensitivity in the 61 SM samples confirmed as malignant was 45.9% (19 BRAF, nine RAS and one TERT). Finally, all 69 LB-FNA classified as M were malignant with an overall mutation rate of 78.3% (54/69). In this category, all mutated cases except one were BRAF V600E (53/54, 98.1%), emphasizing the key role of this oncogene in driving development of PTC. Interestingly, in this group, TERT promoter mutations were associated with BRAF mutation in five cases (7.2%), indicating more aggressive tumours. We then focused on cases presenting with an indeterminate cytology (AUS/FN/SM, n = 257), where this molecular testing might be helpful to better stratify patients for surgery. Importantly we observed that being mutated for this simple molecular panel was still strongly associated with malignancy (P <.0001). Significances of individual BRAF, RAS and TERT status in this indeterminate cytology group are shown in Table 1. Sensitivity of our routine molecular panel to detect malignancy was 47.9% (45/94) and specificity 89.0% (145/163). Positive predictive values (PPV) were respectively 55.0% (11/20), 35.7% (5/14) and 100% (28/28) in AUS, FN and SM categories. Importantly, in those patients actually addressed for surgery according to the Bethesda system guidelines, negative predictive values were 85.4% (35/41) in AUS and 90.0% (99/110) in FN. In detail, RAS mutations significantly predicted cancer (P =.0041). BRAF mutants were only associated with PTC histology whereas N/HRAS mutations were found in both PTC and FTC (Table 1). Cases of NIFTP were only mutated for RAS (33.3%) but not for BRAF or TERT. Of note, TERT promoter mutations were more frequently associated with FTC histology (22.2% vs 3.3% in PTC; P =.0460). Repartition of BRAF, RAS and TERT point mutations in this indeterminate cytology group is shown in Figure 2. One patient with LB-FNA classified as AUS showed a BRAF K601E mutation but no cancer on surgical specimen, whereas the three other patients harbouring this same rare BRAF mutation were all diagnosed with a follicular variant of papillary thyroid carcinoma (one classified as FN and two as SM; Table 2). Contrarily the 70 LB- FNA harbouring a BRAF V600E mutation were all confirmed as malignant. So BRAF K601E mutation should be considered carefully. As previously described NRAS was the most mutated RAS family member mutated in thyroid cancers (see Figure 2). Focusing on TERT promoter mutations in our cohort (n = 326), we identified nine mutated cases (three FN, one SM and five M) all confirmed to be malignant. These included seven papillary carcinomas (four classical PTC, one follicular variant of PTC and two tall cell variant of PTC), and two H urthle cell follicular carcinomas. An example of mutated LB-FNA classified as FN and histology of its paired surgical specimen diagnosed as a follicular carcinoma is shown in Figure S1. Clinicopathological factors for these patients are summarised in Table 3. In two cases classified as FN, TERT mutation was not associated with any BRAF or N/HRAS mutations, and in one case classified as FN harbouring a BRAF K601E mutation together with a C228T TERT promoter mutation was also confirmed to be malignant after surgery. The six remaining cases were associated with a classical BRAF V600E mutation. According to recent studies, TERT promoter mutations were associated with bad prognostic factors: extra thyroid extension (ETE, 4/9) and ptnm stage pt3 (5/9). Unfortunately, follow-up was too short (<2 years in half of the patients) to study impact of these mutations on overall and recurrence free survival. 4 DISCUSSION FIGURE 1 Mutation frequency across the Bethesda categories. Diagnosis of malignancy was confirmed by histology on paired surgical specimens. Population size in each group is indicated (n). Chi-square tests were used to compare frequency between benign and malignant cases Primary goal of LB-FNA in thyroid nodules is to exclude thyroid malignancy to avoid unnecessary surgical thyroidectomies. LB-FNA is a sensitive method to diagnose thyroid nodules, but, in about 20%- 30% of cases, cytology is indeterminate and surgery becomes the only possibility to differentiate benign from malignant lesions. LB- FNA were shown to be particularly suitable for use of ancillary methods such as immunohistochemistry and molecular testing 9,22-24

4 4 DECAUSSIN-PETRUCCI ET AL. TABLE 1 Molecular profiling of thyroid nodules with indeterminate cytology (n = 257) BRAF status NRAS/HRAS status TERT status Total mutated n = 257 n Mutated P value n Mutated P value n Mutated P value n Mutated P value Malignant cytology No (n = 163) (0.6%) < (10.9%) (11.0%) <.0001 Yes (n = 94) (21.3%) (25.3%) 77 4 a (5.2%) (47.9%) Malignant histology (n = 94) PTC (26.3%) (26.0%) ns 60 2 a (3.3%) (51.3%) ns NIFTP (33.3%) (33.3%) FTC (11.1%) 9 2 a (22.2%) 9 2 (22.2%) PTC, papillary thyroid carcinoma; NIFTP, non-invasive follicular thyroid neoplasm with papillary-like nuclear features; FTC, follicular thyroid carcinoma. a When samples are mutated for both TERT and BRAF/NRAS, they account only for 1 in the Total mutated column. TABLE 2 mutations Clinicopathological factors in patients with BRAF K601E Patient sample THD307 THD092 THD058 THD258 Bethesda 3 (AUS) 4 (FN) 5 (SM) 5 (SM) FIGURE 2 Repartition of BRAF, RAS and TERT point mutations in the indeterminate cytology group (n = 258) and comparison between benign (green) and malignant (red) cases Recently, molecular markers were proposed to impact on preoperative decision 9 prompting the ATA to recommend incorporating molecular testing in the algorithms used to manage patients with indeterminate thyroid nodules. 12,25,26 However, implementation of these tests in clinical routine may be difficult. 27 By studying a simple molecular panel (BRAF, NRAS, HRAS and TERT) on DNA extracted from residual material from preoperative thyroid LB-FNA, we successfully characterized a large cohort of 326 cases, demonstrating that this straightforward workflow could easily be implement in clinics. The main advantages of our approach are its cost-effectiveness compared to GEC test and next-generation sequencing (NGS) panel. Its integration in routine clinical practice could be performed with no additional dedicated FNA nor frozen sample needed for molecular testing. As expected, sensitivity of our routine molecular panel to detect malignancy in indeterminate cytology was only 47.9% (Table 1) compared to 94% for the GEC test or 91% for ThyroSeq v2. However, its specificity was comparable to the NGS panel (89.0% vs 92.1%) but much better than the Afirma GEC test (only 17% of confirmed malignancy in GEC-suspicious nodules in an independent cohort) Importantly, negative predictive values were 85.4% in AUS and 90.0% (98/109) in FN, demonstrating that it could help avoiding unnecessary surgery in these two Sex m f f m Age (y) Size (mm) Malignant No Yes Yes Yes Histology Oncocytic follicular adenoma FVPTC FVPTC FVPTC Vascular invasion No No No No ETE No No No Yes TNM NC pt2n0 pt2n0 pt3nx Recurrence No No No No BRAF status K601E K601E K601E K601E TERT status WT C228T WT WT ETE, extra thyroidal extension; AUS, atypia of undetermined significance; FN, follicular neoplasm; SM, suspicious of malignancy; FVPTC, follicular variant of papillary thyroid carcinoma; K601E, BRAF p.lys601glu; C228T, TERT promoter c-124c>t; C250T, TERT promoter c-146c>t; m, male; f, female. indeterminate categories. Our results are in line with study by Shrestha et al (on dedicated samples) showing that performance of simple molecular panels are not significantly different from the NGS panel in the AUS group. 31 Overall mutation rate was relatively low in our cohort from west Europe (35.6%) compared to previous reports. This might have two explanations. First, the panel we studied was restricted to DNA mutations and therefore excluded PAX8/PPARg and RET/PTC rearrangements. Of note, difficulty to extract RNA of sufficient quality from non-dedicated LB-FNA was already pointed out. 32 Second, as previously reported, the most frequent variant of PTC in our region of France is the follicular variant of PTC, known to have less molecular mutations than classical PTC. 19 In our cohort, BRAF V600E mutation was exclusively associated with PTC. RAS mutations were less specific than BRAF V600E with 17 out of 41 RAS mutated cases found to be benign by histology.

5 DECAUSSIN-PETRUCCI ET AL. 5 TABLE 3 Clinicopathological factors in patients harbouring a TERT promoter mutation Patient sample THD586 THD606 THD092 THD421 THD066 THD411 THD454 THD535 THD618 Bethesda 4 (FN) 4 (FN) 4 (FN) 5 (SM) 6 (M) 6 (M) 6 (M) 6 (M) 6 (M) Sex m f f m f f f m m Age (y) Size (mm) Malignant Yes Yes Yes Yes Yes Yes Yes Yes Yes Histology Oncocytic minimally invasive FTC Oncocytic angioinvasive FTC FVPTC Classic PTC Classic PTC Classic PTC Tall cell PTC Tall cell PTC Classic PTC Vascular invasion No Yes No Yes No No No Yes No ETE No No No No Yes No Yes Yes Yes ptnm pt3nx pt2n0 pt2n0 pt2nx pt4an1 pt1bn0 pt3n1a pt3nx pt4an1b BRAF status WT WT K601E V600E V600E V600E V600E V600E V600E N/HRAS status WT WT WT WT WT WT WT WT WT TERT status C228T C228T C228T C250T C228T C228T C228T C228T C250T FVPTC, follicular variant of papillary thyroid carcinoma; PTC, papillary thyroid carcinoma; FTC, follicular thyroid carcinoma; ETE, extra thyroidal extension; FN, follicular neoplasm; SM, suspicious of malignancy; M, malignant; m, male; f, female; WT, wild type; V600E, BRAF p.val600glu; K601E, BRAF p.lys601glu; C228T, TERT promoter c-124c>t; C250T, TERT promoter c-146c>t. However, RAS mutants were still significantly associated with malignancy. Therefore, RAS testing, when positive, might be useful in AUS and FN categories, where risk of malignancy is low, and decision between observation and surgery is difficult. 33 Screening of TERT promoter mutations in thyroid LB-FNA has been poorly studied and never in a west European population. We found 5.5% of malignant cases mutated for TERT, in line with Liu and Xing. 15 In their North American cohort (n = 308) they identified 7% of mutated LB-FNA, all associated with malignancy. Mutation rate was higher in a recently published Korean cohort of 242 preoperative LB-FNA, going up to 16.5%. All mutated cases were confirmed as malignant. 20 Of note, results of the initial Bethesda cytology were not described in any of these studies and might explain differences in frequency (only 69 LB-FNA classified as M in our cohort). We identified TERT promoter mutations in seven papillary carcinomas including two tall cell variants, and two H urthle cell follicular carcinomas. Importantly these mutations detected in LB-FNA were always associated with malignancy even when isolated, suggesting that testing TERT might be of interest in patients with indeterminate cytology. Moreover, TERT mutations were associated with aggressive features such as extrathyroidal extension and high stages. Therefore, similar to BRAF V600Emutated patients, we propose that these TERT-mutated patients should be referred for surgery. Together, these data show that TERT promoter mutation is a rare event but a strong predictor of malignancy that could improve diagnosis accuracy in LB-FNA classified as AUS/FLUS or FN/SFN. and TERT is a strong predictor of malignancy regardless of the cytological classification according to the Bethesda system. Given the high negative predictive values of this molecular panel in patients with indeterminate cytology, we propose to use this strategy to help clinicians and surgeons decision to operate patients classified as AUS or FN. ACKNOWLEDGMENTS We would like to thanks Celine Michaux, Florence Geiguer and Florence Morin for their excellent technical support. CONFLICT OF INTEREST The authors declared that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported. AUTHOR CONTRIBUTIONS MDP, FD and JL: study design, results analysis, manuscript draft; MDP and LD: cytology and histology data; FD and JL: molecular biology data; VL, MLD and JCL: LB-FNA and surgical specimens; FBC: manuscript reviewing. ORCID J. Lopez 5 CONCLUSIONS This study demonstrates that molecular profiling can be routinely performed in a clinical laboratory on LB-FNA. Testing BRAF, N/HRAS REFERENCES 1. McLeod DSA, Sawka AM, Cooper DS. Controversies in primary treatment of low-risk papillary thyroid cancer. Lancet. 2013;381:

6 6 DECAUSSIN-PETRUCCI ET AL. 2. Agretti P, Niccolai F, Rago T, et al. BRAF mutation analysis in thyroid nodules with indeterminate cytology: our experience on surgical management of patients with thyroid nodules from an area of borderline iodine deficiency. J Endocrinol Invest. 2014;37: Bongiovanni M, Spitale A, Faquin WC, Mazzucchelli L, Baloch ZW. The Bethesda System for reporting thyroid cytopathology: a metaanalysis. Acta Cytol. 2012;56: Ohori NP, Schoedel KE. Variability in the atypia of undetermined significance/follicular lesion of undetermined significance diagnosis in the Bethesda System for reporting thyroid cytopathology: sources and recommendations. Acta Cytol. 2011;55: Cooper DS, Doherty GM, Haugen BR, et al. Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid. 2006;16: Crippa S, Mazzucchelli L, Cibas ES, Ali SZ. The Bethesda System for reporting thyroid fine-needle aspiration specimens. Am J Clin Pathol. 2010;134: ; author reply Nikiforov YE, Ohori NP, Hodak SP, et al. Impact of mutational testing on the diagnosis and management of patients with cytologically indeterminate thyroid nodules: a prospective analysis of 1056 FNA samples. J Clin Endocrinol Metab. 2011;96: Nikiforov YE, Nikiforova MN. Molecular genetics and diagnosis of thyroid cancer. Nat Rev Endocrinol. 2011;7: Bongiovanni M, Trimboli P, Rossi ED, Fadda G, Nobile A, Giovanella L. Diagnosis of endocrine disease: high-yield thyroid fine-needle aspiration cytology: an update focused on ancillary techniques improving its accuracy. Eur J Endocrinol. 2016;174:R53-R Alexander EK, Kennedy GC, Baloch ZW, et al. Preoperative diagnosis of benign thyroid nodules with indeterminate cytology. N Engl J Med. 2012;367: Nikiforov YE, Carty SE, Chiosea SI, et al. Highly accurate diagnosis of cancer in thyroid nodules with follicular neoplasm/suspicious for a follicular neoplasm cytology by ThyroSeq v2 next-generation sequencing assay. Cancer. 2014;120: Haugen BR, Alexander EK, Bible KC, et al American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid. 2016;26: Kim TH, Kim Y-E, Ahn S, et al. TERT promoter mutations and longterm survival in patients with thyroid cancer. Endocr Relat Cancer. 2016;23: Liu X, Qu S, Liu R, et al. TERT promoter mutations and their association with BRAF V600E mutation and aggressive clinicopathological characteristics of thyroid cancer. J Clin Endocrinol Metab. 2014;99: E1130-E Liu R, Xing M. Diagnostic and prognostic TERT promoter mutations in thyroid fine-needle aspiration biopsy. Endocr Relat Cancer. 2014;21: Xing M, Liu R, Liu X, et al. BRAF V600E and TERT promoter mutations cooperatively identify the most aggressive papillary thyroid cancer with highest recurrence. J Clin Oncol. 2014;32: Landa I, Ganly I, Chan TA, et al. Frequent somatic TERT promoter mutations in thyroid cancer: higher prevalence in advanced forms of the disease. J Clin Endocrinol Metab. 2013;98:E1562-E Song YS, Lim JA, Choi H, et al. Prognostic effects of TERT promoter mutations are enhanced by coexistence with BRAF or RAS mutations and strengthen the risk prediction by the ATA or TNM staging system in differentiated thyroid cancer patients. Cancer. 2016;122: Sassolas G, Hafdi-Nejjari Z, Ferraro A, et al. Oncogenic alterations in papillary thyroid cancers of young patients. Thyroid. 2012;22: Lee SE, Hwang TS, Choi Y-L, et al. Prognostic significance of TERT promoter mutations in papillary thyroid carcinomas in a BRAF (V600E) mutation-prevalent population. Thyroid:. 2016;26: Nikiforov YE, Seethala RR, Tallini G, et al. Nomenclature revision for encapsulated follicular variant of papillary thyroid carcinoma: a paradigm shift to reduce overtreatment of indolent tumors. JAMA Oncol. 2016;2: Cochand-Priollet B, Dahan H, Laloi-Michelin M, et al. Immunocytochemistry with cytokeratin 19 and anti-human mesothelial cell antibody (HBME1) increases the diagnostic accuracy of thyroid fineneedle aspirations: preliminary report of 150 liquid-based fine-needle aspirations with histological control. Thyroid. 2011;21: Cochand-Priollet B. Thyroid cytopathology: how far can we go? Cytopathology. 2014;25: Rossi ED, Fadda G, Schmitt F. The nightmare of indeterminate follicular proliferations: when liquid-based cytology and ancillary techniques are not a moon landing but a realistic plan. Acta Cytol. 2014;58: Nikiforov YE. Role of molecular markers in thyroid nodule management: then and now. Endocr Pract. 2017;23: Zhang M, Lin O. Molecular testing of thyroid nodules: a review of current available tests for fine-needle aspiration specimens. Arch Pathol Lab Med. 2016;140: Eszlinger M, Bohme K, Ullmann M, et al. Evaluation of a two-year routine application of molecular testing of thyroid fine-needle aspirations using a seven-gene panel in a primary referral setting in Germany. Thyroid. 2017;27: Harrell RM, Bimston DN. Surgical utility of Afirma: effects of high cancer prevalence and oncocytic cell types in patients with indeterminate thyroid cytology. Endocr Pract. 2014;20: Nikiforov YE, Carty SE, Chiosea SI, et al. Impact of the multi-gene thyroseq next-generation sequencing assay on cancer diagnosis in thyroid nodules with atypia of undetermined significance/follicular lesion of undetermined significance cytology. Thyroid. 2015;25: McIver B, Castro MR, Morris JC, et al. An independent study of a gene expression classifier (Afirma) in the evaluation of cytologically indeterminate thyroid nodules. J Clin Endocrinol Metab. 2014;99: Shrestha RT, Evasovich MR, Amin K, et al. Correlation between histological diagnosis and mutational panel testing of thyroid nodules: a two-year institutional experience. Thyroid. 2016;26: Kim Y, Choi KR, Chae MJ, et al. Stability of DNA, RNA, cytomorphology, and immunoantigenicity in residual ThinPrep specimens. APMIS. 2013;121: Ferris RL, Baloch Z, Bernet V, et al. American thyroid association statement on surgical application of molecular profiling for thyroid nodules: current impact on perioperative decision making. Thyroid. 2015;25: SUPPORTING INFORMATION Additional Supporting Information may be found online in the supporting information tab for this article. How to cite this article: Decaussin-Petrucci M, Descotes F, Depaepe L, et al. Molecular testing of BRAF, RAS and TERT on thyroid FNAs with indeterminate cytology improves diagnostic accuracy. Cytopathology. 2017;00:1-6.