SURGICAL UTILITY OF AFIRMA: EFFECTS OF HIGH CANCER PREVALENCE AND ONCOCYTIC CELL TYPES IN PATIENTS WITH INDETERMINATE THYROID CYTOLOGY

Transcription

1 ENDOCRINE PRACTICE Rapid Electronic Article in Press Rapid Electronic Articles in Press are preprinted manuscripts that have been reviewed and accepted for publication, but have yet to be edited, typeset and finalized. This version of the manuscript will be replaced with the final, published version after it has been published in the print edition of the journal. The final, published version may differ from this proof. Original Article EP13330.OR SURGICAL UTILITY OF AFIRMA: EFFECTS OF HIGH CANCER PREVALENCE AND ONCOCYTIC CELL TYPES IN PATIENTS WITH INDETERMINATE THYROID CYTOLOGY Running Title: Surgical Utility of the Afirma Gene Expression Classifier R. Mack Harrell, MD, FACP, FACE, ECNU 1, David N. Bimston, MD, FACS, FSSO 1 From the: 1 Nova Southeastern University College of Osteopathic Medicine Davie, Florida Address correspondence to: R. Mack Harrell, MD, FACP, FACE, ECNU 3287 NW 56 th Street; Boca Raton, FL Rmharrell1@aol.com

2 Abstract Objective: The Afirma GEC (GEC) molecular marker assay was developed for the purpose of improving surgical decision-making with indeterminate fine needle aspiration (FNA) biopsies of thyroid nodules. In this paper, we analyze the performance of the GEC over 27 months in a community hospital-based thyroid surgery practice. Methods: We began using GEC and Thyroid Cytopathology Partners (TCP) exclusively for thyroid FNA analysis in 1/2011, shortly after the Afirma GEC first became commercially available. In this paper, we focus on patients with indeterminate FNA results and the outcomes of GEC analysis, with particular attention to the calculation of the negative predictive value of the Afirma test. Results: We performed 645 FNA s in 519 patients over 27 months. 58 of the FNA s (9%) were read as indeterminate, with 36 of these classified as suspicious by GEC (62%), 20 characterized as GEC benign (34%) and 2 determined to be inadequate due to low mrna content. Of the 36 suspicious GEC patients, 30 underwent thyroidectomy and 21 of the 30 had malignant final pathology. Of the 20 benign GEC patients, 5 underwent thyroid surgery and 2 were discovered to have malignancies. Negative predictive value for the Afirma GEC in our practice environment was 89.6%. Conclusion: In a practice with a high incidence of thyroid cancer in patients with indeterminate FNA s (33% for our practice), the negative predictive value of the Afirma GEC test may not be as robust as suggested in the literature to date. KEY WORDS: Afirma GEC, fine needle aspiration biopsy of thyroid, indeterminate thyroid biopsy, thyroid cancer, molecular markers for thyroid cancer, Hurthle cells, oncocytes

3 Abbreviations: GEC = Afirma GEC; FNA = fine needle aspiration; TCP = Thyroid Cytopathology Partners ; NPV = negative predictive value; TCP = Thyroid Cytopathology Partners ; AACE = American Association of Clinical Endocrinologists; AUS-FLUS = atypia of undetermined significance or follicular lesion of undetermined significance; FN = follicular neoplasm. Introduction Since the landmark report of Kimura et al in 2003 linking the BRAF mutation with papillary thyroid cancer (1), endocrinologists and endocrine surgeons have held great expectations that one day genetic analysis of thyroid fine needle aspiration biopsy specimens might decrease the number of unnecessary surgeries performed for patients with cytologically indeterminate thyroid nodules. While the presence of the V600E BRAF mutation proved to be highly predictive of papillary cancer, the fact that 40-60% of patients with a tissue diagnosis of papillary cancer do not carry a BRAF mutation has limited the diagnostic applicability of this genetic test (2). Researchers have subsequently attempted to identify other genetic mutations and translocations that could predict thyroid cancer. One after another, mutations in RAS and translocations in RET/PTC and PAX8/PPAR gamma have proven to be useful when present in patient cytological samples, but unhelpful for patients whose thyroid cancers do not exhibit known genetic alterations (2). In 2010, Chudova et al took a different approach by analyzing the amplified transcriptome (mrna) of fine needle aspiration biopsy specimens (FNA s) from patients undergoing thyroid surgery, with the intention of developing a gene expression test that could predict benignity. Mathematical analysis led to the development of a 142 gene cdna Affymetrix cassette (Afirma GEC) capable of separating benign from suspicious FNA specimens in a small independent cohort of prospectively collected samples. Initial statistical analysis was promising with a benign GEC result yielding a negative predictive value (NPV) of 96% (3). Further testing led to the Alexander et al paper in 2012 that further validated the Afirma GEC in FNA s from 265 patients with cytologically indeterminate nodules, with the authors reporting an overall NPV of 94% in patients with pre-operative negative GEC s (4).

4 However, negative predictive value has an inverse relationship with the overall prevalence of thyroid cancer in the indeterminate cytology group studied (5). We reasoned that the incidence of thyroid cancer in patient groups with indeterminate thyroid cytology might vary according to geographic region and type of thyroid cancer practice. Therefore, we chose to evaluate our South Florida (Memorial Center for Integrative Endocrine Surgery) Afirma GEC experience over the past 27 months and to calculate our own practice-specific NPV. In addition, we examined cytological and pathological idiosyncrasies in our patient group that might potentially affect GEC performance. Methods From 1/2011 through 4/2013, all thyroid FNA cytology specimens from our South Florida endocrine surgery practice were sent to Thyroid Cytopathology Partners (TCP) of Austin, Texas along with an Afirma washout specimen of RNA preserved in FNA-protect. Patients were selected for thyroid nodule FNA based on American Association of Clinical Endocrinologists (AACE) Thyroid Nodule Guideline criteria, and all biopsies were performed under General Electric Logiq E9 or S8 ultrasound guidance using a parallel-to-the-beam technique. Each nodule was sampled with three to five 25 gauge needle passes and slides were fixed in 95% ethanol and dried for transport. Each needle s contents were washed into FNA-protect after initial use for slide creation. After cytologic characterization at TCP, FNA-protect washes from all of our 58 indeterminate biopsies were sent to Veracyte, Inc. in South San Francisco, California for Afirma GEC testing. The nomenclature for indeterminate FNA result classification is divided into two groups including (1) atypia of undetermined significance or follicular lesion of undetermined significance (AUS-FLUS), either follicular or Hurthle Cell predominant, and (2) follicular neoplasm (FN), either follicular or Hurthle Cell predominant. Throughout the text of this report, the terms Hurthle Cell and oncocyte are used interchangeably. FNA s read as suspicious for malignancy were not included in our indeterminate FNA analysis because Veracyte has excluded them from Afirma GEC testing since the summer of Over the course of the study, only four FNA s were interpreted as suspicious for malignancy by Thyroid Cytology Partners and all proved to be cancers at surgery.

5 All surgery was performed by one surgeon (DNB) and final pathologic reports were rendered by the Memorial Health Systems Department of Pathology in consultation with Dr. Virginia LiVolsi at the University of Pennsylvania, Dr. Jennifer Hunt at the University of Arkansas and the Pathology Staff of the University of Florida in Gainesville. Negative predictive value was estimated using the standard equation for negative predictive value and our data was presented using a conventional 2 x 2 table with pre-surgical Afirma GEC results, benign and suspicious, plotted versus clinically significant surgical pathologic diagnoses, benign and malignant (Table 1) (5). Four subcentimeter papillary microcarcinomas, contralateral from the actual nodule biopsied, were not considered to be clinically significant malignancies. Results Nodule and Patient Characteristics We performed 645 FNA s in 519 patients over 27 months. The 58 biopsies read as indeterminate by Thyroid Cytopathology Partners ranged in size from 0.6 to 5 cm, averaging 1.5 cm. We sampled six sub-centimeter nodules that were found to have indeterminate cytology. These sub-centimeter nodules were biopsied because of suspicious ultrasound findings or strong family histories of thyroid cancer and constituted only 10.3% of the entire indeterminate group. The 58 indeterminate nodules represented 9% of the total FNA s sent to Thyroid Cytopathology Partners over the time frame of the study. 36 of these indeterminate biopsies were classified as suspicious by Afirma GEC (62%), 20 were characterized as GEC benign (34%) and 2 were determined to be inadequate due to low mrna content (3%). The 58 patients whose thyroid FNA s were sent for Afirma GEC analysis averaged 53 years of age and were predominantly female (43 females in the group or 76%). The ethnicity of the group was largely Caucasian (67%), with Hispanic (16%), African American (14%) and Asian (3%) minority representation. Surgical Results

6 Of the 36 suspicious GEC patients, 30 (83%) underwent thyroidectomy. Twentyone of the 30 had malignant final pathology. Four of the 21 patients with malignant final pathologic diagnoses had isolated co-incidental sub-centimeter papillary microcarcinomas in the contralateral thyroid lobe from their biopsied Afirma GEC suspicious nodules (the false positive GEC lesions ranged from 1.0 to 1.6 cm in size). Of the GEC suspicious patients who did not undergo surgery, 5 refused surgical intervention and one is in the process of completing chemotherapy for breast cancer prior to her thyroidectomy. Of the 20 benign GEC patients, 5 underwent thyroid surgery and 2 were discovered to have malignancies. The false negative results in these two malignant thyroid nodules were likely due to sampling error, with the first nodule being a 0.64 cm papillary cancer, and the second a 2.8 cm cystic papillary cancer. Surgery was performed in the first patient because a frankly malignant aspirate in the contralateral thyroid lobe and in the second case because of a positive family history and recent nodule growth. All GEC benign patients were encouraged to return for yearly follow-up with sequential neck ultrasound evaluation. One of the two patients with non-diagnostic GEC s (inadequate RNA) underwent lobectomy for a benign thyroid cyst. Calculation of NPV In an effort to construct a best case scenario for calculation of the negative predictive value of the Afirma GEC in our practice, we excluded the four incidental and remote papillary microcarcinomas and assumed a minimum cancer prevalence of 17 cases in the Afirma suspicious group and 2 cases in the Afirma benign group for a total of 19 cancers in 58 patients. Thus our minimum cancer prevalence in the indeterminate FNA group was 33% (Table 1). This adjustment assumes that no more relevant thyroid cancers will be found in the 22 patients who have yet to undergo a thyroid surgical procedure. In addition, we counted the false negative GEC 0.64 cm papillary microcarcinoma as a true negative (technically the nodule was too small to biopsy and a large malignant lesion in the contralateral thyroid lobe drove our decisionmaking) and calculated the best case NPV for Afirma in our practice to be 89.6% (Table 1, Figure 1). If we assume that a minimum of two more relevant cancers would be uncovered at final pathologic evaluation (consistent with the cancer incidence in the already operated group) from the six as yet unoperated, suspicious GEC patients, our indeterminate FNA cancer prevalence rises to 36% and NPV falls to 88.3% FNA Cellular Typology Cytologically, 21 out of 58 (41%) of our indeterminate FNA s showed prominent Hurthle Cell populations.

7 In the Afirma GEC benign group, only 2 of 20 aspirates were noted by the TCP cytologists to contain significant Hurthle Cell (oncocytic) populations. The 18 nononcocytic FNA s included 16 AUS-FLUS lesions featuring follicular cells with mild atypia and 2 follicular neoplasms. In contrast, in the Afirma GEC suspicious group, 19 out of 36 FNA s (53%) showed oncocytic predominance, with 12 Hurthle Cell dominant AUS-FLUS FNA s and 7 oncocytic follicular neoplasms (FN). Note that biopsies demonstrating Hurthle Cells in proximity to copious colloid and lymphocytes are not included in this discussion because Thyroid Cytopathology Partners interprets such FNA s as benign by Bethesda criteria. Pathologic Correlates 36 surgeries were performed in our 58 patients with indeterminate FNA s (30 on GEC suspicious FNA patients, 5 on GEC benign FNA patients and 1 on a nondiagnostic GEC patient with insufficient RNA) (Table 2). Ten oncocytic tumors were diagnosed at pathological evaluation (10/36 patients or 28%) including 7 Hurthle Cell adenomas, 2 multinodular goiters with Hurthle Cell metaplasia and 1 Hurthle Cell carcinoma (Table 2-black fill, column 4). If we include the four contralateral thyroid lobe papillary microcarcinoma patients as false positive GEC s (Table 2- red text), 9 of 13 false positive Afirma GEC patients (69%) were subsequently shown to have benign oncocytic pathology. Thyroidectomy for patients with GEC suspicious oncocytic FNA s demonstrated a clinically relevant cancer on final pathology in 6 out of 17 surgeries (35%- Table 2). Once again, for the purposes of this analysis, we classified four contralateral thyroid lobe incidental papillary microcarcinomas as benign (i.e. not clinically significant), based on the fact that these lesions are routinely present in 6-36% of the healthy adult population (6, 7). Using these classification criteria, a patient with a suspicious, oncocytic Afirma GEC result has roughly the same risk for harboring a clinically relevant thyroid cancer as a patient with a Bethesda 4 FNA result. Thus, for a patient with an oncocytic indeterminate FNA, Afirma GEC adds little information to enhance surgical decision making. Thyroidectomy in patients with Afirma GEC suspicious, non-oncocytic, indeterminate FNA s produced a clinically relevant cancer on final pathology in 10 out 13 surgeries (77%- Table 2). Thus, the GEC suspicious, non-oncocytic, indeterminate FNA group appeared to be enriched with relevant thyroid cancer. None of the five surgeries performed on patients with benign Afirma GEC s produced an oncocytic tumor; nor did the surgery on the patient with a nondiagnostic GEC (Table 2- blue and gray fill).

8 Discussion In this paper, we analyze 27 successive months of Veracyte Afirma GEC use in our endocrine surgical practice and compare our GEC negative predictive value to that reported by Alexander et al in a much larger study involving 49 sites (mostly nonuniversity endocrinology practices) in 26 states over roughly the same duration. Like the participants in the Alexander et al study, we used Thyroid Cytopathology Partners, Inc. of Austin, Texas (TCP) for all cytologic interpretation, but unlike Alexander et al, we insisted on ethanol-fixation and air-drying of all our slides prior to transport. We did not use Thin-Prep liquid technology. Because the Pap technique requires wet staining, our slides were not examined with Papanicoau staining and this could have resulted in Bethesda classification differences. In addition, we did not use 1-2 dedicated passes for Afirma GEC RNA collection, choosing instead to wash all our needle passes into FNA-protect solution after creating a slide for cytology. This technique did not appear to compromise the quality of the RNA specimens we provided, as only 2 of 58 Afirma GEC s (3%) were found to be RNA inadequate. Indeterminate Group Cancer Prevalence and NPV While the sex distribution and the average age of our Afirma patient population are similar to that described in the Alexander et al study, our cancer prevalence and final pathologic diagnosis distribution is different (4). The cancer prevalence in our indeterminate group was at least 33% (probably closer to 36% once all Afirma GEC suspicious surgeries are complete), while the calculated cancer prevalence (with suspicious for malignancy cytology patients excluded) in the Alexander et al paper was almost 9% lower at 24.3%. The increased prevalence of thyroid cancer in our indeterminate group leads to a significant reduction in the negative predictive value of the Afirma GEC in our practice population. When cancer prevalence is plotted on a Cancer Prevalence vs NPV curve constructed from our practice sensitivity and specificity data (curve fitting algorithm supplied by Dr. Bryan McIver of the Moffitt Cancer Center), our calculated NPV falls to 89.6% (Table 1, Figure 1). Thus, for our practice, more than one in ten of our benign GEC Afirma patients are likely to be misclassified. Although our current data is limited by the fact that 15 out of 20 patients in our Afirma GEC benign group have not undergone surgery, our experience still aptly demonstrates that the higher

9 the prevalence of thyroid cancer in the tested population, the more likely that Afirma GEC will misclassify patients with malignant disease. Oncocytic Case-Mix In another divergence from the Alexander et al data, our indeterminate FNA patient group demonstrated a considerably higher percentage of oncocytic final pathologic diagnoses (28%) than did the multicenter trial (12%) (4). Although the reason for our plethora of Hurthle Cell dominant final diagnoses is unclear, it is apparent that Afirma technology disproportionately distributes oncocytic indeterminate lesions into the suspicious GEC category (19 out of 21 oncocytic biopsies were classified as Afirma GEC suspicious) and that most of these Hurthle Cell lesions are benign on final pathologic evaluation (9 of 13 of the GEC false positive nodules contained dominant oncocytic cell populations on final pathology, including 7 Hurthle Cell adenomas and 2 multinodular goiters with Hurthle Cell metaplasia (Table 2). It is possible that the high concentration of oncocytic aspirates in our indeterminate specimens (21/58 or 36.2%) could have affected Afirma performance. 90% of our oncocytic aspirates (19 out of 21) were subsequently read as GEC suspicious. Alexander et al (4) do not provide a breakdown of their indeterminate cytologies into oncocytic and non-oncocytic subsets, so we cannot directly compare our data to theirs. However, the fact that our prevalence of oncocytic final pathology was more than double theirs (28% vs 12%), suggests that that our indeterminate cytologic cell type distribution was likely different from theirs. Historically, Hurthle Cell adenomas and carcinomas have been cytologically indistinguishable. Recent genomic deep sequencing studies provide evidence that Hurthle Cell carcinomas are genetically unique thyroid malignancies, distinct from papillary thyroid and follicular thyroid cancers (8). Given the infrequency of Hurthle Cell malignancies, it is possible that the Afirma GEC was not adequately trained to differentiate benign from malignant oncocytic indeterminate FNA s. Although novel genetic Hurthle Cell markers have been described (8, 9), no commercially available genetic test has been shown to reliably distinguish benign from malignant oncocytic tumors. Summary Our 27 month study of the Afirma Genetic Expression Classifier reveals that the negative predictive value for this thyroid surgery-preventing test diminishes considerably in populations where the indeterminate cytology group (excluding suspicious for

10 malignancy patients) contains more than 1 in 4 patients with thyroid cancer. Our study is limited by a relatively small sample size and by the fact that 22 of our patients with indeterminate FNA s (17 in the Afirma GEC benign group) have not undergone surgery. With such a limited sample size, a small number of misclassified cancers by Afirma testing can dramatically alter the sensitivity and specificity calculations for the GEC test and secondarily change negative predictive value calculations. However, we encourage all thyroidologists and thyroid surgeons who use Afirma GEC technology to calculate or estimate practice-specific thyroid cancer prevalences in their indeterminate cytology patient groups. Using such cancer prevalence data, a practice-based re-estimation of the negative predictive value of a benign Afirma GEC test may prove to be informative. In endocrine surgical practices like ours, the negative predictive value of a benign Afirma GEC test may not be so robust as Alexander et al suggest (4). The problem of Hurthle Cell-rich thyroid aspirates that do not contain copious colloid or cellular evidence for chronic thyroiditis remains a vexing one. Commercially available genetic technology seems incapable of distinguishing benign from malignant Hurthle Cell lesions. Traditionally, oncocytic indeterminate cytology patients have been sent to surgery for diagnosis. Over our first 27 months of Afirma GEC use, 19 of 21 patients with indeterminate oncocytic cytologies were deemed suspicious by Afirma and sent for surgical diagnosis. Thus, in our practice, Afirma GEC added little information to the surgical selection process in patients with Hurthle Cell dominant cytopathology.

11 References 1. Kimura ET, Nikiforova MN, Zhu Zhaowen, et al. High Prevalence of BRAF Mutations in Thyroid Cancer: Genetic Evidence for Constitutive Activation of the RET/PTC-RAS-BRAF Signaling Pathway in Papillary Thyroid Carcinoma. Cancer Research, 2003; 63: Nikiforova MN and Nikiforov YE. Molecular Diagnostics and Predictors in Thyroid Cancer. Thyroid, 2009; 19: Chudova D, Wilde JI, Wang, ET. Molecular Classification of Thyroid Nodules Using the High-Dimensionality Genomic Data. Clin Endocrinol Metab. 2010; 95: Alexander EK, Giulia CK, Zubair WB. Preoperative Diagnosis of Benign Thyroid Nodules with Indeterminate Cytology. N Engl J Med; 367: Fletcher, RH. Clinical Epidemiology: The Essentials. 4 th ed. Philadephia, PA: Lippincott Williams &Wilkins, Lang W, Borrusch H, Bauer L. Occult Carcinomas of the Thyroid: Evaluation of Sequential Autopsies. Am J Clin Pathol. 1988:90: Harach H, Franssila K, Wasenius V. Occult Papillary Carcinoma of the Thyroid: A Normal Finding in Finland. Cancer. 1985;3: Ganly I, Filho JR, Eng, S. Genomic Dissection of Hurthle Cell Carcinoma Reveals a Unique Class of Thyroid Malignancy. J Clin Endocrinol Metab. 2013;98 (early release). 9. Cerutti, JM, Oler, G, Delcelo R. PVALB, a New Hurthle Adenoma Diagnostic Marker Identified through Gene Expression. J Clin Endocrinol Metab. 2011;96:E151-E1

12 Figure Captions Figure 1. Plot of NPV/PPV versus Prevalence of Malignancy in the Afirma GEC Tested Group. (NPV line in blue, PPV in red)-graph courtesy of Dr. Bryan McIver and the Moffitt Cancer Center. The best case scenario thyroid cancer prevalence in our indeterminate cytology group was 33% with a calculated Afirma GEC negative predictive value (NPV) of 89.6% represented by the white arrow.

13 Figure 1 NPV / PPV NPV 89.6% Cancer Prevalence - 33% Prevalence of malignancy Figure 1. Plot of NPV/PPV versus Prevalence of Malignancy in the Afirma GEC Tested Group. (NPV line in blue, PPV in red)-graph courtesy of Dr. Bryan McIver and the Moffitt Cancer Center. Our best case scenario thyroid cancer prevalence in our indeterminate cytology group was 33% with a calculated Afirma GEC negative predictive value (NPV) of 89.6% represented by the white arrow.

14 2 x 2 Table of Afirma GEC Results Vs Final Patholoigc Results (With Sensitivity, Specificity, NPV and PPV Calculations) Final Pathologic Diagnosis Malignant Benign Afirma Suspicious GEC Benign 1 4 N Sensitvity Specificity NPV PPV Max cancer prevalence (assuming 51.4% prev in indeterminates) Min cancer prevalence (no further cancer in 22 unoperated pts) Table 1: Afirma GEC Results Vs Final Pathologic Diagnoses Sensitivity and specificity calculations are in bold black. Negative predictive value (NPV) and positive predictive value (PPV) calculations are in italics.two assumptions are depicted: The first in red italics assumes that the cancer prevalence in the 22 unoperated patients will be identical to that seen in the 35 operated patients-51.4%. The second assumption (best case scenario) assumes that there will be no further cancers detected in the 22 unoperated patients (black italics)- 33%

15 Table 2. Cytologic and Pathologic Correlates in 36 Patients Undergoing Surgery After Afirma GEC Pt # FNA Bethesda Category GEC Dx Pathology Cancer Stage 1 3 Benign Pap Ca- multifocal Stage Benign MNG 3 3 Benign MNG 4 4 Benign Pap Ca Stage Benign MNG 6 3 Inadequate RNA Benign Cyst 7 4-HC Suspicious HC Carcinoma Stage HC Suspicious HC Adenoma 9 3-HC Suspicious MNG with HC changes 10 3-HC Suspicious Pap Ca- multifocal Stage 1* 11 3-HC Suspicious HC Adenoma 12 4-HC Suspicious HC Adenoma 13 3-HC Suspicious CLT w/ micro Pap 14 4-HC Suspicious HC Adenoma 15 3 Suspicious Pap Ca- multifocal Stage HC Suspicious Pap Ca Stage HC Suspicious HC Adenoma w/ micro Pap 18 3 Suspicious Pap Ca Stage Suspicious Pap Ca- Follicular Variant Stage Suspicious Pap Ca Stage Suspicious Pap Ca Stage 4A 22 3 Suspicious Pap Ca Stage 4A 23 3 Suspicious Pap Ca- multifocal Stage 1* 24 3 Suspicious Pap Ca Stage 1* 25 3 Suspicious Follicular nodule 26 3-HC Suspicious Pap Ca Stage HC Suspicious HC Adenoma w/ micro Pap 28 4-HC Suspicious MNG with HC changes 29 3 Suspicious Pap Ca- Follicular Variant Stage 1* 30 3 Suspicious Pap Ca- Follicular Variant Stage Suspicious CLT w/ micro Pap 32 4-HC Suspicious HC Adenoma 33 3-HC Suspicious Pap Ca- multifocal Stage 1* 34 3 Suspicious Benign reactive tissue 35 3-HC Suspicious Pap Ca- multifocal Stage HC Susp/Medullary Medullary Carcinoma Stage 4A

16 Legend to Table 2 Color coding: Abbreviations: Blue Cells- Benign Afirma GEC Gray Cell- Inadequate RNA GEC Yellow Cells- Suspicious GEC Green Cells- Suspicious Medullary GEC White Cells- False Positive Afirma GEC s Black Cells- Oncocytic Final Pathologic Diagnosis Red Text- Incidental opposite lobe papillary microcarcinoma CLT- Chronic Lymphocytic Thyroiditis HC Adenoma- Hurthle Cell Adenoma HC Carcinoma- Hurthle Cell Carcinoma MNG- Multinodular Goiter Pap Ca- Papillary Carcinoma *All Stage 1 thyroid cancers in column 5 are T1N0M0