Journal Club American Journal of Clinical Pathology, January 2011. Dr.

Nidhi Jais

The Predictive Value of the FineNeedle Aspiration Diagnosis Suspicious for a Follicular Neoplasm, Hrthle Cell Type in Patients With Hashimoto Thyroiditis
Michael H. Roh, MD, PhD,1* Vickie Y. Jo, MD,2* Edward B. Stelow, MD,2 William C. Faquin, MD,3. Kelly H. Zou, PhD,4 Erik K. Alexander, MD,5 P. Reed Larsen, MD,5 Ellen Marqusee, MD,5, Carol B. Benson, MD,6 Mary C. Frates, MD,6 Atul Gawande, MD,7 Francis D. Moore Jr, MD,7 and Edmund S. Cibas, MD8

Introduction
Hrthle cell neoplasms are relatively rare thyroid neoplasms of follicular cell origin. Hrthle cell carcinoma (HCC) constitutes only about 3% of all thyroid carcinomas. Earlier it was thought that HCCs behave more aggressively than follicular carcinomas but later no statistically significant difference in the rate of local recurrence, metastasis, or patient survival was found. Thus, HCC has been reclassified by the WHO as an oncocytic variant of follicular carcinoma.

Introduction
The detection of thyroid nodules has increased lately and FNA represents a crucial element in their initial evaluation and has improved preoperative assessment significantly. A thyroid aspirate that is cellular and consists exclusively of Hrthle cells and lacks the nuclear features of papillary carcinoma raises the possibility of a Hrthle cell neoplasm.

Introduction
The lack of chronic inflammation or colloid in the background and the presence of transgressing blood vessels have been shown to be significantly associated with Hrthle cell neoplasms on histologic follow-up. Most significant, these cytologic findings raise the possibility of HCC. Nevertheless, Hrthle cell adenomas outnumber HCCs; hence, the specificity of the suspicious for a follicular neoplasm, Hrthle cell type (SFNHCT) diagnosis for malignancy is low.

Introduction
The Hrthle cellonly pattern, reported as FNHCT or as SFNHCT can be seen in some nonneoplastic conditions1. Multinodular goiter 2. Lymphocytic (Hashimoto) thyroiditis (HT).

Introduction
Hyperplastic nodules in the setting of multinodular goiter and HT can vary in the extent of Hrthle cell metaplasia and occasionally give rise to aspirates so rich in Hrthle cells that the aspirate is interpreted as SFNHCT. As a result, non-neoplastic nodules represent a significant proportion (up to 40%) of the histologic outcome after an interpretation of FNHCT/ SFNHCT.

Introduction
The positive predictive value (PPV) for malignancy of the FNA interpretation SFNHCT has been reported to be 15% to 45%. Although, the overall PPV of SFNHCT has been established, no studies to date have examined it in patients with HT. Given the high frequency of non-neoplastic Hrthle cell proliferations in patients with HT, they hypothesized that the PPV for malignancy in patients with HT might be lower than in patients without HT.

Aims and Objectives

To determine whether the PPV for malignancy of SFNHCT is different in patients with and without a history of HT.

Review of literature
Carcangiu ML, Bianchi S, Savino D, et al. A clinicopathologic study of 153 cases of thyroid follicular Hurthle cell tumor (HCT) is presented. These neoplasms were divided into three categories on the basis of presence and degree of capsular and vascular invasion, pattern of growth, nuclear atypia, and necrosis. The categories were benign (90), indeterminate (35), and malignant (28). All the tumors classified histologically as benign or indeterminate behaved in a clinically benign fashion. Of the 28 tumors classified histologically as malignant, 20 had a clinically malignant behavior. It is concluded that histologic criteria alone can distinguish benign from malignant HCT and that clinical or pathologic feature cannot predict behavior among the malignant tumors.

Review of literature
Gosain AK, Clark OH studied the records of patients with Hrthle cell neoplasms from 1943 to 1982. Of 84 patients, 71 had Hrthle cell adenomas (HCAs), 9 had Hrthle cell change in chronic thyroiditis, and 4 had Hrthle cell carcinomas. Coexisting papillary thyroid carcinoma occurred in 3 patients with HCA. 12 patients with HCAs had multiple lesions, 5 of which were bilateral. There were no recurrences or deaths among the patients with benign Hrthle cell tumors. Thus, patients with HCA had a benign course, and histologic examination results accurately reflected malignant potential.

Materials and Methods

After approval by the institutional review boards electronic pathology databases at 3 institutions were searched to identify patients who underwent thyroid FNA between 1992 and 2007 with cytologic findings that were suspicious for a Hrthle cell neoplasm. Brigham and Womens Hospital, Boston, MA; Massachusetts General Hospital, Boston; University of Virginia ,Charlottesville

Materials and Methods

Patients with histologic follow-up after a partial or total thyroidectomy were identified, and the diagnoses for the previously aspirated nodules were noted after the correlation of gross descriptions with ultrasound guided FNA notes. The sizes of these nodules and the diagnoses of neoplasia (adenoma and carcinoma) and hyperplasia, specifically, were tabulated based on the original histologic interpretations recorded in the surgical pathology reports.

Materials and Methods

HCCs, papillary thyroid carcinomas (PTCs), and poorly differentiated carcinomas were diagnosed based on
presence of venous and/or transcapsular invasion nuclear features insular growth pattern in poorly differentiated carcinomas

HCC

PTC

PDC

Materials and Methods

Thyroid FNAs were processed differently at the 3 institutions.
Brigham and Womens Hospital
liquid-based preparations optional cell blocks

Massachusetts General Hospital

ethanol-fixed, Pap smears ethanol-fixed, H&E-stained smears; air-dried, Romanowskystained smears; and/or ThinPrep slides.

University of Virginia
air-dried, rapid Romanowsky stained smears ethanol-fixed, Papanicolaoustained smears with or without ThinPrep slides.

Materials and Methods

The electronic medical record was searched to determine whether the patient had a history of HT at the time of initial examination of the thyroid nodule before the thyroid FNA and subsequent surgery - examining the medical history in clinical notes.  whether receiving levothyroxine replacement therapy and/or was clinically hypothyroid.  titers for antibodies against thyroid peroxidase (TPO).  findings on thyroid ultrasonography.

Materials and Methods

Finally, the histopathologic findings were examined to determine which patients had pathologic evidence of lymphocytic thyroiditis in follow-up thyroidectomy specimens. Cases without surgical follow-up were excluded from the study.

Materials and Methods

A patient was classified as having HT if one of the following criteria was satisfied: (1) a clinical history of HT, with elevated anti-TPO titers and documentation of levothyroxine replacement therapy; (2) clinical history of HT, with a diffusely heterogeneous thyroid gland on ultrasound, and documentation of levothyroxine replacement therapy; (3) diffuse lymphocytic thyroiditis on histologic examination of surgically resected thyroid tissue.

Materials and Methods

The PPV of the cytologic interpretation of SFNHCT for neoplasia and that for malignancy were calculated for patients with and without HT based on the follow-up histologic diagnoses as the gold standard. The Fisher exact test was used to assess statistical difference between these 2 groups. The Wilcoxon rank sum test was used to assess the statistical difference in nodule sizes between the 2 groups. Statistical significance was determined if the P value was less than 0.05.

Results
Between 1992 and 2007, a total of 401 patients with cytology suspicious for a Hrthle cell neoplasm were identified. Of these patients, surgical follow-up was available for 287(71.6%). Review of clinical patient encounter notes, anti-TPO titers, thyroid ultrasonography, and surgical specimens revealed that 21 of 287 (7.3%) patients met our criteria for HT.

Results
19 of 21 had a clinically documented history of HT.

14 required levothyroxine replacement therapy 10 had ultrasonographic findings suggestive of HT

8 had elevations in anti-TPO titers

Results
One had ultrasonographic evidence of HT, an elevated anti-TPO titer, and histologic evidence in the resected thyroid gland. In the other patient, ultrasonographic and laboratory data were not available; however, examination of the surgically resected thyroid gland revealed diffuse lymphocytic thyroiditis consistent with HT.

Results
Of the 287 patients with histologic follow-up, the size of the aspirated nodules was available for 275 (95.8%). The mean SD nodule size for the overall study population was 2.6 1.4 cm (range, 0.5-10 cm). There was no significant difference between the mean size of the nodules in the HT and non-HT cohorts.

Results

Results
Hurthe cell carcinoma, 41 Papillary carcinoma, 2

Poorly differentiate d carcinoma, 26

Distribution of malignant neoplasm in 69 of 287 cases.

Results
In the follow-up surgical specimens, the previously aspirated nodules represented neoplastic nodules (adenomas and carcinomas) in 9 of 21 (42.9%) and 183 of 266 (68.8%) patients with and without HT, respectively. This difference was statistically significant (P = .016).

Results
Neoplastic nodules- 9 Patients with Carcinomas- 2

HT (21)

Patients without HT(266)

Neoplastic nodules-183 Carcinomas-67

Results
The PPV for the cytologic interpretation of SFNHCT specifically for a malignant neoplasm in the non-HT cohort was 25.2% (67 of 266): 39 HCCs, 26 PTCs, and 2 poorly differentiated carcinomas. In contrast, the PPV for malignancy in the HT cohort was 9.5% (2 of 21); both malignancies were HCCs. Although the PPV of SFNHCT for malignancy was lower in the HT group, the trend did not reach statistical significance (P = .081).

Results

Malignant neoplasms in patients without Hashimoto thyroiditis (HT). A, C, and E, representative examples of thyroid FNA cases with cytologic findings suspicious for a follicular neoplasm, Hrthle cell type for 3 patients without HT

Results

B, A Hcc was confirmed by the presence of transcapsular invasion D, An oncocytic variant of PTC was identified; one portion of the nodule consisted of oncocytic follicular cells (inset) which merged with an area exhibiting more pronounced nuclear atypia with chromatin clearing, nuclear membrane irregularities, and nuclear grooves F, Poorly differentiated carcinoma composed of crowded insular nests of follicular cells with oncocytic cytoplasm

Discussion
Hyperplastic nodules in the setting of HT can occasionally yield an aspirate composed exclusively of Hrthle cells. The Hrthle cellonly pattern, interpreted as SFNHCT, raises the possibility of a neoplasm, specifically Hrthle cell adenoma and, more significant, an HCC. To date, however, the significance of a Hrthle cell only aspirate in the context of HT, especially the PPV of SFNHCT for neoplasia and malignancy in patients with HT, has not been thoroughly investigated.

Discussion
During a 15-year period, they identified 21 patients with HT with cytologic features of SFNHCT. In comparison, the remaining 266 patients with cytologic findings of SFNHCT did not have HT. Next, they determined the frequency of neoplasms diagnosed on histologic follow up and compared the PPV for neoplasia in patients with and without HT.

Discussion
They found that among the patients with HT, fewer than half of the aspirated nodules were neoplasms. This proportion was significantly less than that seen in the non-HT cohort of 266 patients, in which 68.8% of the aspirated nodules were neoplasms. 68.8% in nonHT cohort 42.9% HT cohort

Discussion
If fewer than half of thyroid nodules in patients with HT cytologically interpreted as SFNHCT are neoplasms, what is the PPV for malignancy in the setting of HT? Overall, for the entire study population, the PPV for malignancy was 24%. The majority of the malignancies were HCCs; however, a significant number of PTCs were diagnosed. Prior reports corroborate this finding and collectively these data testify to the difficulty in interpreting nuclear atypia of Hrthle cell neoplasms on FNA.

Discussion
After stratifying the cases based on the presence or absence of a history of HT, we found only 2 malignancies in the cohort of 21 patients with HT. In contrast, 25.2% of the patients without HT were found to harbour malignancies on histologic follow-up.

Discussion
It has been reported that HT could be a risk factor for the development of PTC. None of the patients with HT in the study, however, was diagnosed with PTC after surgery. This is likely due to specific analysis of thyroid FNA cases in which the cytologic findings were SFNHCT as opposed to suspicious for papillary thyroid carcinoma or positive for papillary thyroid carcinoma.

Discussion
The lower PPV for malignancy of SFNHCT in the HT cohort compared with the non-HT cohort did not reach statistical significance, largely owing to the small patient cohort fulfilling our criteria for HT, despite pooling data from 3 large medical centers. The diagnosis of SFNHCT is a relatively infrequent diagnosis; an interpretation of follicular neoplasm/suspicious for follicular neoplasm is given almost 3 times more frequently than FNHCT/SFNHCT. 401 patients with thyroid cytologic features of SFNHCT were identified.

Discussion
Criteria used for interpreting thyroid aspirates as SFNHCT: A cellular specimen composed exclusively of Hrthle cells, arranged as isolated cells or clusters, with little or no colloid. Plasma cells, lymphocytes, and lymphohistiocytic aggregates were absent or very sparse.

Discussion
Because the PPV for malignancy of SFNHCT in the setting of HT is not zero and because the lower PPV for malignancy did not reach statistical significance, they do not advocate that a benign interpretation be made in a patient with HT for an aspirate that would be normally interpreted as SFNHCT. The PPV for malignancy of thyroid aspirates interpreted as atypia of undetermined significance (AUS) is estimated to be 5% to 15%, in contrast with the PPV of SFNHCT or suspicious for follicular neoplasm, which range from 15% to 45% and 15% to 30%,respectively.

Discussion
As the PPV of SFNHCT for malignancy in the HT cohort was approximately 10%, they propose that a thyroid FNA obtained from a patient known to have HT that exhibits predominantly Hrthle cells can be interpreted as AUS or SFNHCT. An SFNHCT interpretation more accurately reflects the cytologic pattern but overstates the risk of malignancy. If interpreted as SFNHCT, the interpretation can be accompanied by an educational note that the risk of malignancy is lower than that of SFNHCT in general.

Discussion
On the other hand, an AUS interpretation more accurately reflects the malignancy risk of the nodule A diagnosis of AUS is typically managed by a repeated thyroid FNA. Although the interpretation of AUS would allow for greater clinical flexibility, for example, a period of watchful waiting rather than referral for surgical lobectomy.

Discussion
This is illustrated by the 2 patients with HT with HCC identified on follow-up. For these patients, delaying surgical management could have been harmful because HCC has a propensity for lymphovascular space invasion and spread to regional lymph nodes. Their study results have revealed a discordance between the predicted and the actual risk of malignancy associated with an SFNHCT cytologic interpretation in patients with HT.

Discussion
The best way to report cases that fit this scenario is likely to remain controversial. This underscores the importance of further investigation into the pathology and molecular biology of HCC. In addition, more experience with the application of the Bethesda System to this scenario may lead to a consensus in the future.

Happy Doctors Day

References
1. DeLellis RA, Lloyd RV, Heitz PU, et al, eds. IARC WHO Classification of Tumours: Pathology and Genetics of Tumoursof Endocrine Organs. Lyon, France: IARC Press; 2004. 2. Thompson NW, Dunn EL, Batsakis JG, et al. Hrthle cell lesions of thyroidgland. Surg Gynecol Obstet.1974;139:555-560. 3. Hundahl SA, Fleming ID, Fremgen AM, et al. A National Cancer Data Base report on 53,856 cases of thyroid carcinoma treated in the US, 1985-1995. Cancer.1998;83:2638-2648. 4. Carcangiu ML, Bianchi S, Savino D, et al. Follicular Hrthle cell tumors of the thyroid gland. Cancer. 1991;68:1944-1953. 5. Gosain AK, Clark OH. Hrthle cell neoplasms: malignant potential. Arch Surg. 1984;119:515-519. 6. Evans HL, Vassilopoulou-Sellin R. Follicular and Hrthle cell carcinomas of the thyroid: a comparative study. Am J Surg Pathol. 1998;22:1512-1520. 7. Segev DL, Saji M, Phillips GS, et al. Polymerase chain reactionbased microsatellite polymorphism analysis of follicular and Hrthle cell neoplasms of the thyroid. J Clin Endocrinol Metab. 1998;83:2036-2042.

References
8. Troncone G, Iaccarino A, Russo M, et al. Accumulation of p27(kip1) is associated with cyclin D3 overexpression in the oxyphilic (Hrthle cell) variant of follicular thyroid carcinoma. J Clin Pathol. 2007;60:377-381. 9. French CA, Alexander EK, Cibas ES, et al. Genetic and biological subgroups of low-stage follicular thyroid cancer.Am J Pathol. 2003;162:1053-1060. 10. Nikiforova MN, Lynch RA, Biddinger PW, et al. RAS point mutations and PAX8-PPAR gamma rearrangement in thyroid tumors: evidence for distinct molecular pathways in thyroid follicular carcinoma. J Clin Endocrinol Metab. 2003;88:2318-2326. 11. Yassa L, Cibas ES, Benson CB, et al. Long-term assessment of a multidisciplinary approach to thyroid nodule diagnostic evaluation. Cancer. 2007;111:508-516. 12. Elliott DD, Pitman MB, Bloom L, et al. Fine-needle aspiration biopsy of Hrthle cell lesions of the thyroid gland:a cytomorphologic study of 139 cases with statistical analysis.Cancer. 2006;108:102-109. 13. Amrikachi M, Ramzy I, Rubenfeld S, et al. Accuracy of fine-needle aspiration of thyroid: a review of 6226 cases and correlation with surgical or clinical outcome. Arch Pathol Lab Med. 2001;125:484-488.

References
14. Gharib H, Goellner JR, Johnson DA. Fine-needle aspiration cytology of the thyroid: a 12-year experience with 11,000 biopsies. Clin Lab Med. 1993;13:699-709. 16. Pu RT, Yang J, Wasserman PG, et al. Does Hrthle cell lesion/neoplasm predict malignancy more than follicular lesion/neoplasm on thyroid fine-needle aspiration? Diagn Cytopathol. 2006;34:330-334. 17. Baloch ZW, LiVolsi VA, Asa SL, et al. Diagnostic 2008;36:425-437. 18. Montone KT, Baloch ZW, LiVolsi VA. The thyroid Hrthle (oncocytic) cell and its associated pathologic conditions: asurgical pathology and cytopathology review. Arch Pathol Lab Med. 2008;132:1241-1250. 19. Gharib H, Goellner JR. Fine-needle aspiration biopsy of the thyroid: an appraisal. Ann Intern Med. 1993;118:282-289. 20. Repplinger D, Bargren A, Zhang YW, et al. Is Hashimotos thyroiditis a risk factor for papillary thyroid cancer? J Surg Res. 2008;150:49-52. 21. Cibas ES, Ducatman BS. Cytology: Diagnostic Principles and Clinical Correlates. 3rd ed. Philadelphia, PA: Saunders; 2009. 22. Yang J, Schnadig V, Logrono R, et al. Fine-needle aspiration of thyroid nodules: a study of 4703 patients with histologic and clinical correlations. Cancer. 2007;111:306315.

Ultrasound It is difficult to reliably sonographically differenciate Hashimoto thyroiditis from other thyroid pathology. Ultrasound features can be variable depending of the severity and phase of disease1,5. a diffusely enlarged thyroid gland with a heterogeneous echotexture is a common sonographic presentation 6. the presence of hypoechoic micronodules ( 1 - 6 mm) with a surroundning echogenic rim is also considered to have a relatively high positive predictive value

hypoechoic thyroid gland 2) coarse echotexture of the gland 3) fine linear echoes within the thyroid parenchyma s/o fibrosis 4) Color doppler imaging reveals augmentation of the vascularity of the thyroid gland. These ultrasound images are diagnostic of Hashimoto's thyroiditis. On thyroid scan the pattern of uptake may be homogeneous to spotty and uneven; scans may mimic hot or cold nodules. Note, the radioactive iodine uptake is often elevated in patients with Hashimoto's thyroiditis at the time of TSH elevation. A thyroid scan is generally not necessary in these patients. The thyroid ultrasound appearance is characteristic but not diagnostic. The pattern is diffusely heterogeneous, often somewhat hyperechoic. To the uninitiated these are often misinterpreted as multinodular glands although discrete nodules are not seen.

AUS- Scattered clusters of atypical follicular cells, scant colloid Few benign follicular cells, scant colloid

I. Nondiagnostic or unsatisfactory Cyst fluid only Virtually acellular specimen Other (obscuring blood, clotting artifact, etc) II. Benign Consistent with a benign follicular nodule (includes adenomatoid nodule, colloid nodule, etc) Consistent with lymphocytic (Hashimoto) thyroiditis in the proper clinical context Consistent with granulomatous (subacute) thyroiditis III. Atypia of undetermined significance/follicular lesion of undetermined significance IV. Follicular neoplasm/"suspicious" for follicular neoplasm Specify if Hrthle cell type

V. Suspicious for malignancy Suspicious for papillary carcinoma Suspicious for medullary carcinoma Suspicious for metastatic carcinoma Suspicious for lymphoma VI. Malignant Papillary thyroid carcinoma Poorly differentiated carcinoma Medullary thyroid carcinoma Undifferentiated (anaplastic) carcinoma Squamous cell carcinoma Carcinoma with mixed features Metastatic

multiple small irregular echopenic and linear echogenic areas and irregularity of the thyroid borders. Ultrasound It is difficult to reliably sonographically differenciate Hashimoto thyroiditis from other thyroid pathology. Ultrasound features can be variable depending of the severity and phase of disease1,5. a diffusely enlarged thyroid gland with a heterogeneous echotexture is a common sonographic presentation 6. the presence of hypoechoic micronodules ( 1 - 6 mm) with a surroundning echogenic rim is also considered to have a relatively high positive predictive value 3,4. Scintigraphy Radioactive iodine early stages : may show increased uptake late stages : single or multiple areas of reduced uptake (cold spots).