Barriers to the Recognition of Medullary Thyroid Carcinoma on FNA: Implications Relevant to the New American Thyroid Association Guidelines

Transcription

1 Barriers to the Recognition of Medullary Thyroid Carcinoma on FNA: Implications Relevant to the New American Thyroid Association Guidelines Kathryn S. Dyhdalo, MD and Deborah J. Chute, MD BACKGROUND: The 2016 American Thyroid Association guidelines recommend multiple endocrine neoplasia testing and evaluation for pheochromocytoma before thyroidectomy after a thyroid fine-needle aspiration biopsy (FNA) is positive for medullary thyroid carcinoma (MTC). In the current study, the authors examined the reasons why FNA was unable to definitively diagnose MTC preoperatively, with attention to morphologic patterns that can be misleading. METHODS: Cases of MTC diagnosed on thyroid surgical resection for which there was a prior FNA and slides available for review were included (28 cases). Clinicopathologic data were collected. Slides were reviewed for numerous features blinded to the original FNA interpretation. Morphologic features were compared between concordant cases (20 cases) ( positive for MTC or suspicious for MTC ) and discordant cases (4 cases) (any other interpretation). Three cases of microscopic MTC (measuring <1 cm) were excluded from statistical analysis, as was 1 case of sampling error (benign thyroid tissue). RESULTS: Nine men and 19 women were diagnosed with MTC. Four patients ultimately were diagnosed with multiple endocrine neoplasia, and 1 had bilateral pheochromocytomas. At the time of surgical excision, the mean tumor size was 2.3 cm (range, cm). Review of morphologic features demonstrated that the discrepant cases were significantly more likely to have limiting factors (air-drying artifact/excess blood), a cohesive pattern, or to lack plasmacytoid morphology. None of the discordant cases had pseudoinclusions or amyloid (finding was not statistically significant). CONCLUSIONS: The majority (86%) of thyroid FNAs from patients with MTC are concordant (positive/suspicious for MTC). Patterns of failure include sampling error and limited typical morphologic features, particularly a lack of plasmacytoid morphology and cellular dyshesion. A high level of suspicion for MTC is critical to ensure patients receive appropriate preoperative testing. Cancer Cytopathol 2018;126: VC 2018 American Cancer Society. KEY WORDS: American Thyroid Association (ATA); cytology; fine-needle aspiration (FNA); medullary thyroid carcinoma (MTC); multiple endocrine neoplasia (MEN) syndrome; pitfalls; thyroid. INTRODUCTION Fine-needle aspiration (FNA) is a common and useful sampling technique with which to evaluate thyroid nodules. It is minimally invasive, more common than surgical biopsy, and carries fewer complications for patients. 1 Medullary thyroid carcinoma (MTC) is an aggressive thyroid tumor that can be reliably diagnosed on FNA. 2 The newly revised treatment guidelines from the American Thyroid Association (ATA) 3 for MTC suggest numerous steps before proceeding to total thyroidectomy: workup for RET germline mutations, evaluation for pheochromocytoma (if RET mutated), and adrenalectomy if pheochromocytoma is identified. This decision pathway is summarized in Figure 1. Therefore, it is becoming more important to identify and correctly diagnose MTC on FNA samples. However, the diagnosis of FNA samples is not always straightforward. Limited sample, variable tumor morphology, poor slide preparation, and variable experience and comfort level of the cytopathologist all Corresponding author: Deborah J. Chute, MD, Department of Pathology, Pathology and Laboratory Medicine Institute, Cleveland Clinic, 9500 Euclid Ave, L25, Cleveland, OH 44195; chuted@ccf.org Department of Pathology, Pathology and Laboratory Medicine Institute, Cleveland Clinic, Cleveland, Ohio Received: December 5, 2017; Revised: February 26, 2018; Accepted: February 27, 2018 Published online March 26, 2018 in Wiley Online Library (wileyonlinelibrary.com) DOI: /cncy.21990, wileyonlinelibrary.com Cancer Cytopathology June

2 Figure 1. Summary of the decision pathway. - indicates negative; 1, positive; FNA, fine-needle aspiration; MTC, medullary thyroid carcinoma. can influence the diagnosis of MTC. 4 6 In this study, we retrospectively examined the reasons why FNA biopsy is unable to recognize MTC preoperatively in a manner sufficient for clinical management, with specific attention to morphologic patterns that can be misleading. This has particular implications given the new ATA guidelines to prevent a pheochromocytoma crisis in the operating room. MATERIALS AND METHODS The current study was approved by the institutional review board of the Cleveland Clinic. The pathology database (CoPath) was searched for cases of MTC diagnosed at the time of surgical resection from January 1, 1980, to April 6, 2016, and for which there was a prior ipsilateral FNA biopsy in the institution s system. Surgical resections of nonthyroid origin (eg, lymph node) were excluded. Of 129 cases of MTC in the system during this time, 61 cases from 61 unique patients were identified that met the inclusion criteria. Information collected from the surgical pathology report included the size of the MTC and other pathologic findings present. Information from the original FNA report was collected, including patient age at the time of FNA, 398 Cancer Cytopathology June 2018

3 Barriers to MTC Recognition on FNA/Dyhdalo and Chute original interpretation, and any ancillary testing performed. The Bethesda System for Reporting Thyroid Cytopathology 2 was implemented in 2010 at the study institution. Cases from before that date often used non-bethesda terminology; these cases were placed into the best-fit category for simplicity. For example, one sample originally was diagnosed as suspicious for malignancy with a listed differential diagnosis of suspicious for follicular neoplasm (SFN) versus follicular variant of papillary thyroid carcinoma (PTC); this case was listed in the suspicious category. Additional information was collected from the electronic medical record, including patient demographics, family history of multiple endocrine neoplasia (MEN) or MTC, results of RET testing, MEN diagnosis, results of pheochromocytoma workup, and other clinical follow-up. In the study institution, pheochromocytoma workup includes laboratory analysis for plasma catecholamines (epinephrine, norepinephrine, and dopamine) and plasma metanephrines (metanephrine and normetanephrine). In addition, testing for RET germline mutations is initiated by clinicians and involves send-out testing of peripheral blood to the Mayo Clinic for germline sequencing. In one unusual case, the patient had germline testing performed (with negative results) and the clinical team later submitted resected tumor tissue for FoundationOne testing (a next-generation sequencing-based assay, Foundation Medicine, Mass). Of the 61 cases that met the inclusion criteria, 28 had FNA slides available. These cases were reviewed for a variety of features including adequacy according to the current Bethesda System for Reporting Thyroid Cytopathology, cellularity, limiting factors (excess blood, air-drying artifact), material in a cell block, and morphologic criteria commonly observed in aspirates of MTC, 2,7,8 including the predominant pattern present (spindled, epithelioid, and/or oncocytic), the presence of pseudoinclusions, plasmacytoid cytoplasm, cellular dyshesion, salt-and-pepper chromatin, multinucleation, and the presence of amyloid. Theslideswerereviewedbybothauthorsatadouble scope, blinded to the original FNA interpretation. Slides for each case consisted of Papanicolaou-stained smears and a ThinPrep slide, and in some cases a hematoxylin and eosin-stained Cellient cell block (if requested by the original pathologist). The cell blocks were created with the Cellient Automated Cell Block System (Cytyc Corporation, Marlborough, Massachusetts) according to the manufacturer s protocol. Immunocytochemical slides, if available, also were reviewed. TABLE 1. Antibody Information Antibody Clone Vendor Dilution Calcitonin Polyclonal Dako 1:8000 CEA Polyclonal Dako 1:2000 Chromogranin DAK-A3 Dako 1:100 Thyroglobulin Polyclonal Dako 1:8000 Abbreviation: CEA, carcinoembryonic antigen. Immunocytochemical stains at the study institution are performed on the Ventana Benchmark Ultra automated immunostainer (Ventana Medical Systems, Tucson, Arizona). Localization of the antigen-antibody complex is achieved using the OptiView DAB IHC Detection Kit (Ventana Medical Systems). All of the stains reviewed were performed on cell blocks at the time of initial diagnosis and incorporated into the original diagnosis. No immunocytochemistry was performed for the purposes of the current study. A summary of the antibodies and laboratory methods can be found in Table 1. Cases definitively diagnosed as MTC at the time of FNAwerecomparedwithcasesinwhichtheFNAwas discordant. The original FNA was considered concordant when the interpretation was positive for MTC or suspicious for MTC and discordant when another diagnosis was rendered. Diagnoses of positive for MTC and suspicious for MTC are considered sufficient for patient workup and clinical intervention to ensure that a medullary carcinoma is not missed. A recent meta-analysis used thesamediagnosticthresholdwhenreviewingtheaccuracy of FNA cytology for the diagnosis of MTC. 9 Statistical Analysis Statistical analysis was performed using a 2-tailed Fisher exact test (GraphPad Software Inc, San Diego, California). P values <.05 were considered statistically significant. RESULTS Retrospective Case Review Clinical data A retrospective case review of 61 patients was performed. There were 23 men and 38 women diagnosed with MTC, with an average age of 57 years (range, years). Six patients had a family or personal history diagnostic or concerning for MEN type 2 (MEN2) syndrome before the FNA: 1 patient had a cousin with MTC, 1 patient had a family history of MEN2, 1 patient had a pheochromocytoma diagnosed Cancer Cytopathology June

4 TABLE 2. Original FNA Results of All 61 Cases Benign AUS SFN Suspicious Positive Other Total Microscopic MTC MTC: 0 PTC: 1 1 cm MTC MTC: 8 Other: 1 MTC: 3 PTC: 3 MTC: 31 Other: Abbreviations: AUS, atypia of uncertain significance; FNA, fine-needle aspiration; MTC, medullary thyroid carcinoma; PTC, papillary thyroid carcinoma; SFN, suspicious for follicular neoplasm. 1 year before their MTC diagnosis, and 3 patients had a concerning personal or family history that was not otherwise elaborated on in the medical record. The remaining patients had no history (45 patients), the history was not documented (9 patients), or the history was of uncertain significance (1 patient diagnosed with Hirschsprung disease as an infant). Nine additional patients ultimately were diagnosed with MEN syndrome after the diagnosis of MTC, either on the basis of RET mutation testing (8 patients) or strong clinical suspicion (1 patient in whom RET was not tested). Five of these patients did not have any personal or family history that was concerning for MEN syndrome before the initial FNA diagnosis. Pathologic features Review of the surgical pathology reports demonstrated the MTCs ranged in size from 0.1 to 7.5 cm in greatest dimension (mean, 2.18 cm; 3 cases had no measurement provided). Ten cases were microscopic MTC (measuring <1 cm). When these cases were excluded, the average size of the MTCs was 2.55 cm. Other pathologic features present included PTC, follicular adenoma, follicular carcinoma of Hurthle cell type, and multinodular goiter with degenerative changes. The FNA interpretation (Table 2) was classified as benign in 2 cases, as atypia of uncertain significance (AUS) in 4 cases, as SFN in 3 cases, as suspicious for malignancy in 10 cases (8 for MTC, 1 for PTC, and 1 suspicious for malignancy [follicular neoplasm vs follicular variant of PTC]), as positive for malignancy in 40 cases (34 for MTC,3forPTC,2casesoflargecell/spindlecellcarcinoma with a differential diagnosis of MTC vs PTC, and 1classifiedascarcinomafavorPTC),andasotherinterpretations in 2 cases (1 case with a limited sample of atypical cells, partially spindled, could not exclude origin from neoplasm and 1 case of atypical cells, favor origin from thyroid neoplasm, with a differential diagnosis including MTC, PTC, and Hurthle cell neoplasm). Ten of the cytology reports (16.4%) noted limiting factors present in the specimen that impeded interpretation: the majority listed low cellularity in the specimen or cell block (9 cases) and 1 case was compromised by air-drying artifact. Twenty-five of the 61 cases had immunocytochemistry performed or attempted on cell blocks, although 6 cases (24%) were limited in some way. Two cases could only support a limited panel (Congo red only), and 4 cases had insufficient positive material with which to render a definitive diagnosis. The most common immunocytochemical stains performed included calcitonin (19 cases), carcinoembryonic antigen (CEA) (13 cases), and chromogranin (12 cases), with 16 cases found to be positive for calcitonin, 12 found to be positive for CEA, and 11 found to be positive for chromogranin. RET testing and pheochromocytoma workup Thirty-six of the 61 patients were tested for a germline RET mutation; 27 showed no germline mutation, 8 demonstrated a mutation, and 1 case was negative for a germline RET mutation although sequencing of the resected tumor tissue showed a variant of uncertain significance (RET D378_G385>E).Thisalterationisanovelmutation. The patient subsequently was treated with vandetanib and achieved a favorable response, as reported in a recent article by Vanden Borre et al. 10 The germline mutations that were identified included 2 cases each of V804M and C611Y and 1 case each of C634Y, C620G, C634R, and C609Y. Thirty-seven patients had workups performed for pheochromocytoma, 2 of which were positive. One of these 2 patients had bilateral pheochromocytomas (this patient also had a C620G RET mutation) and the other patient s pheochromocytoma was discovered on an incidental computed tomography scan performed before their thyroid mass was discovered (RET testing was not performed). The remaining 35 patients had a negative workup for pheochromocytoma. 400 Cancer Cytopathology June 2018

5 Barriers to MTC Recognition on FNA/Dyhdalo and Chute TABLE 3. Original FNA Results of Cases With FNA Slides Available for Review (N 5 28) Benign AUS SFN Suspicious Positive Other Total Microscopic MTC MTC: cm MTC MTC: Other: 2 Abbreviations: AUS, atypia of uncertain significance; FNA, fine-needle aspiration; MTC, medullary thyroid carcinoma; SFN, suspicious for follicular neoplasm. Figure 2. Examples of concordant cases with typical features of medullary thyroid carcinoma, including (A) salt-and-pepper chromatin and (B) dyshesive, multinucleated cells with plasmacytoid morphology (Papanicolaou stain, high-power magnification). Review of Cases With Available FNA Material Clinical data A total of 28 cases had FNA material available for review. There were 9 men and 19 women diagnosed with MTC, with an average age of 60 years (range, years). Three patients had a family or personal history that was diagnostic or concerning for MEN2 syndrome before the FNA: 1 patient had a family history of MEN2 and 2 patients had a concerning personal or family history that was not otherwise elaborated on in the medical record. The remaining 18 patients had no history. Four patients ultimately were diagnosed with MEN syndrome after the diagnosis of MTC; these are the same 4 patients that had RET mutations. Two of these patients did not have any personal or family history that was concerning for MEN syndrome before the initial FNA diagnosis. Pathologic features Review of the surgical pathology reports demonstrated that themtcsrangedinsizefrom0.2to7.5cmingreatest dimension (mean, 2.32 cm). Three cases were microscopic MTC (those measuring <1 cm). When these cases were excluded, the average size was 2.54 cm. Other pathologic features present included multinodular goiter with degenerative changes. The original FNA interpretation (Table 3) was benign in 1 case, AUS in 4 cases, SFN in 1 case, suspicious for malignancy in 5 cases (all cases were deemed suspicious for MTC ), and positive for malignancy in 17 cases (15 for MTC, 1 for large cell carcinoma [MTC vs PTC], and 1 for PTC with MTC mentioned in the differential diagnosis). The case with large cell carcinoma consisted of a moderately cellular sample of malignant epithelioid cells in cohesive clusters. The given differential diagnosis included MTC and PTC, with anaplastic carcinoma less likely but not excluded entirely. Immunocytochemical stains were not performed. Six of the cytology reports (21%) noted low cellularity in the sample that impeded interpretation; no other limitations were recorded. There were some significant differences in the characteristicsoftheaspiratesthatwereconcordantversusdiscordant. Examples of concordant cases with typical MTC features are shown in Figure 2. Examples of discordant cases lacking typical features of MTC are shown in Figure 3. Cases that were discordant (ie, not recognized as MTC atthetimeoffna)weremorelikelytohavelimitedfeaturessuchasairdrying(25%vs15%;p 5.04) and to lack typical morphologic features of MTC such as plasmacytoid cells (25% vs 90%; P 5.02) and dyshesion (25% vs 90%; P 5.02). Cancer Cytopathology June

6 Figure 3. Examples of discordant cases lacking typical features of medullary thyroid carcinoma (MTC). Many samples that were not diagnosed as MTC had atypical morphology, including (A) unremarkable chromatin, (B) cohesive groups, (C) lack of plasmacytoid morphology, and (D) lack of multinucleation (Papanicolaou stain, high-power magnification). Fourteen of the 28 cases had immunocytochemistry performed or attempted on cell blocks, although 3 cases (21%) were noted as limited by low cellularity on the cell block. The most common immunocytochemical stains performed included calcitonin (13 cases), CEA (9 cases), and thyroglobulin (6 cases), with 11 cases found to be positive for calcitonin and 9 cases found to be positive for CEA. Thyroglobulin was negative in 2 cases and noncontributory in 4 cases. In the 28 cases reviewed, there were 3 cases of microscopic MTC ranging from 0.2 to 0.8 cm in greatest dimension. The smallest microscopic MTC was termed AUS at the time of FNA, and the 2 larger cases were termed positive for MTC at the time of the original FNA. Another case consisted of normal thyroid epithelium with abundant colloid on an FNA of the left thyroid lobe; the resection specimen demonstrated a 1.5-cm MTC in the left thyroid lobe and a 0.7-cm colloid nodule in the thyroid isthmus. The assumption in this case was that the cytology specimen sampled the isthmus lesion, rather than the MTC, because the FNA contained colloid and cyst contents with rare benign follicular cells. No features of MTC were apparent on review of the FNA material for this study. These 4 cases were excluded from further analysis due to concern for sampling error. The remaining 24 cases were composed of discordant cytology (4 cases; 3 classified as AUS and 1 as SFN) or concordant cytology (positive or suspicious for MTC) (20 cases; 5 classified as suspicious for MTC and 15 as positive for MTC). A comparison of classic morphologic features between these 2 categories is outlined in Table 4. FNA is an accepted modality for the diagnosis of MTC and is quite sensitive in the hands of an experienced operator. 7,11 However, the variable tumor morphology still can create diagnostic challenges. In the current series of 61 casesofmtcthathadbothapreoperativefnaanda thyroidectomy, MTC was diagnosed correctly in 77% of cases (47 of 61 cases). This number rises to 86% (44 of 51 cases) if the 10 cases of microscopic MTC are excluded. These results are at the high end of the range of concordance reported previously in a recent meta-analysis. 9 RET testing and pheochromocytoma workup Twenty of the 28 patients were tested for a germline RET mutation; 16 demonstrated no germline mutation and 4 were found to have a mutation. The mutations that were 402 Cancer Cytopathology June 2018

7 Barriers to MTC Recognition on FNA/Dyhdalo and Chute TABLE 4. Comparison of Classic Morphologic Features of MTC Present in Discordant Versus Concordant Cases a Discordant Cases N 5 4 Concordant Cases N 5 20 P b BSRTC adequate 4 (1 due to atypia) High cellularity (estimated) Blood limiting Air-drying limiting Predominant pattern 2 spindled 2 epithelioid 2 spindled 6 epithelioid (1 focal, mostly stripped nuclei) 1 oncocytic 3 oncocytic/epithelioid 8 spindled/epithelioid Pseudoinclusions present Plasmacytoid cells Salt-and-pepper chromatin Dyshesive Amyloid Multinucleation Material on cell block N/A Yes: 7 No: 3 N/A: 10 NA Abbreviations: BSRTC, Bethesda System for Reporting Thyroid Cytopathology; MTC, medullary thyroid carcinoma; N/A, not available; NA, not applicable. a Concordance was defined as positive or suspicious for MTC. b Bold type indicates statistical significance. NA identified included 1 case each of C620G, C634R, C609Y, and V804M. A total of 21 patients had workups performed for pheochromocytoma, 1 of which was positive. This patient was found to have bilateral pheochromocytomas (this patient also had a C620G RET mutation). The remaining 20 patients had a negative workup for pheochromocytoma. DISCUSSION The recently updated ATA management guidelines for MTC 3 state that patients who present with newly detected MTC should be screened for evidence of hereditary MTC via physical examination and RET mutational testing; patients with positive screening should be evaluated further for pheochromocytoma and hyperparathyroidism. Given the fact that the majority of clinically worrisome thyroid nodules are evaluated by FNA before surgical treatment, these updated guidelines highlight the importance of a definitive diagnosis of MTC before surgery, given the risk of pheochromocytoma. After reviewing the electronic medical record and cytology cases, we noticed 3 main patterns of discordant FNA interpretation and surgical findings of MTC, with discordant results defined as any result not positive or suspicious for MTC (6 cases total, 4 of which were discordant plus 1 discordant microscopic MTC and 1 sample of normal thyroid). It is important to note that not every discordant case in our database was available for review. The first pattern of discordant FNA interpretation was due to sampling error (2 of 6 cases). This included 1 case of microscopic MTC (<1cm),and 1case inwhichonly normal thyroid tissue was aspirated despite the presence of a concerning lesion on radiologic examination. The second pattern was limited material, resulting in limited morphologic evaluation and/or ancillary testing (2 of 6 cases). A paucicellular cell block and inability to perform confirmatory immunocytochemical testing often were cited as barriers to confident diagnosis. However, it should be noted that many cases with limited material were diagnosed as suspicious for MTC ; although not definitive, this wording would be sufficient to alert the clinicians to the possibility of MTC and initiate the RET/ pheochromocytoma preoperative workup. The third pattern of discrepancy was due to limited morphologic features of MTC being present (6 of 6 cases). The majority of misclassified cases had only 1 or none of the so-called classic morphologic features of medullary carcinoma, including plasmacytoid or spindled cells, dyshesion, multinucleation, and salt-and-pepper chromatin. 2 None of the discordant cases had amyloid or nuclear pseudoinclusions. It is interesting to note that in our case cohort serologic testing was only recommended if a case Cancer Cytopathology June

8 already was considered at least suspicious for MTC. There was a single case diagnosed as positive for malignancy favor PTC; differential diagnosis includes MTC that suggested serology would be helpful. There were no cases for which serology was recommended that did not include MTC in either the line diagnosis or the differential diagnosis. Although it is well known that MTC can exhibit a variety of morphologic patterns in FNA samples, 11,12 the current literature appears to be lacking in the prevalence of specific classic features in each case, and the morphologic features that often are absent or focal and that may lead the pathologist to an incorrect (or less correct, in terms of suspicious for MTC ) diagnosis. One notable pitfall is the nonspecific staining of oncocytic cells with calcitonin, 12 which should prompt caution with regard to overreliance on immunocytochemical stains. Thyroglobulin can be ordered with calcitonin in all oncocytic lesions to avoid this pitfall; a positive thyroglobulin stain would suggest an oncocytic neoplasm rather than a medullary carcinoma. Another notable pitfall stems from samples that combine relatively preserved cellular cohesion with a lack of classic nuclear features of MTC, resulting in a diagnosis of SFN. Aspirates with an oncocytic pattern lacking classic nuclear features may result in an interpretation of SFN, Hurthle cell type. Both are important because they will not trigger preoperative testing and will, at least initially, cause patients to receive inadequate surgical management. The cytologic features of microscopic MTC are even more variable than those of regular MTC. These cases tend to be more cohesive, contain colloid (possibly from sampling of adjacent thyroid tissue), and may not contain plasmacytoid or spindled cells. 5 Extra caution is needed wheninterpretingsmalllesions.inthe3casesofmicroscopic MTC that had slides available for review, all cases had adequate cellularity and an epithelioid cell pattern; none contained amyloid, nuclear pseudoinclusions, or multinucleated cells. Two of the cases had salt-and-pepper chromatin with dyshesive cells; these originally were interpretedasmtc.thecasethatdidnothavethesefeatures was interpreted as AUS. We thought it would be interesting to retroactively apply the new ATA guidelines to these FNA cases and determine how many patients would have undergone RET and pheochromocytoma evaluation before surgery based on the FNA diagnosis. We assumed that any patient with an FNA diagnosis of negative or AUS would not be a surgical candidate based on cytology results alone. A total of 6 patients thus were excluded (5 benign or AUS cases and 1 case with pre-bethesda wording that was unclear regarding the surgical indication [ limited sample of atypical cells, partially spindled, cannot exclude origin from neoplasm. ]). A significant minority of the remaining cases (9 of 55 cases; 16%) would have been missed (ie, the patients had undergone surgery without suspicion or workup for RET/pheochromocytoma). Study limitations included the inability to review slides from every MTC FNA specimen that was performed, and the necessity of a long study period given the rarity of MTC. During the study period, there were changes made in the management of both MTC specificallyandthyroidlesionsingeneral,aswellasintheclassification and reporting of FNA results. 2 The diagnosis of MTC is challenging and requires a high index of suspicion. The recognition of minimal criteria is key to making the diagnosis, with a low threshold to either perform (by immunocytochemistry) or suggest (by serology) further testing. If the new ATA guidelines had been in place during the study period, approximately 16% of the patient population in the current study would have been inappropriately sent for surgery before receiving a RET/pheochromocytoma workup. FUNDING SUPPORT No specific funding was disclosed. CONFLICT OF INTEREST DISCLOSURES The authors made no disclosures. AUTHOR CONTRIBUTIONS Conceptualization: Deborah J. Chute. Datacollection: Kathryn S. Dyhdalo and Deborah J. Chute. Formal analysis:deborah J. Chute. Investigation: Kathryn S. Dyhdalo and Deborah J. Chute. Data curation: Kathryn S. Dyhdalo. Writing original draft: Kathryn S. Dyhdalo. Writing review and editing: Kathryn S. Dyhdalo and Deborah J. Chute. Resources: Kathryn S. Dyhdalo. Supervision: Deborah J. Chute. REFERENCES 1. Hamberger B, Gharib H, Melton LJ 3rd, Goellner JR, Zinsmeister AR. Fine-needle aspiration biopsy of thyroid nodules. Am J Med. 1982;73: Ali SZ, Cibas ES, eds. The Bethesda System for Reporting Thyroid Cytopathology. New York: Springer; Wells SA, Asa SL, Dralle H, et al; American Thyroid Association Guidelines Task Force on Medullary Thyroid Carcinoma. 404 Cancer Cytopathology June 2018

9 Barriers to MTC Recognition on FNA/Dyhdalo and Chute Revised American Thyroid Association guidelines for the management of medullary thyroid carcinoma. Thyroid. 2015;25: Sams SB, Tompkins KD, Mayson S, Raeburn CD, Mehrotra S. Oncocytic variant of medullary thyroid carcinoma; a rare tumor with numerous diagnostic mimics by fine needle aspiration. Diagn Cytopathol. 2017;45: Yang GC, Fried K, Levine PH. Detection of medullary thyroid microcarcinoma using ultrasound-guided fine needle aspiration cytology. Cytopathology. 2013;24: Pusztaszeri MP, Bongiovanni M, Faquin WC. Update on the cytologic and molecular features of medullary thyroid carcinoma. Adv Anat Pathol. 2014;21: Forrest CH, Frost FA, de Boer WB, Spagnolo DV, Whitaker D, Sterrett BF. Medullary carcinoma of the thyroid: accuracy of diagnosis of fine-needle aspiration cytology. Cancer. 1998;84: Green I, Ali SZ, Allen EA, Zakowski MF. A spectrum of cytomorphologic variations in medullary thyroid carcinoma: fine-needle aspiration findings in 19 cases. Cancer. 1997;81: Trimboli P, Treglia G, Guidobaldi L, et al. Detection rate of FNA cytology in medullary thyroid carcinoma: a meta-analysis. Clin Endocrinol (Oxf). 2015;82: Vanden Borre P, Schrock AB, Anderson PM, et al. Pediatric, adolescent, and young adult thyroid carcinoma harbors frequent and diverse targetable genomic alterations, including kinase fusions. Oncologist. 2017;22: Chang TC, Wu SL, Hsiao YL. Medullary thyroid carcinoma: pitfalls in diagnosis by fine needle aspiration cytology and relationship of cytomorphology to RET proto-oncogene mutations. Acta Cytol. 2005;49: 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. Cancer Cytopathology June