—  SPECIALTY CONFERENCE  —

Cytopathology

Case 6 - Hurthle Cell Carcinoma with Focal Anaplastic Carcinoma

Dr. William C. Faquin
Massachusetts General Hospital
Boston, Massachusetts


Click on each slide thumbnail image for an enlarged view
Clinical History:
An 86 year-old woman presents with an 8.0 cm right thyroid mass. An FNAB was performed. (Images 6A-6C)

Cytologic Findings:
The smears, stained with rapid H&E and Papanicolaou stains, are cellular and consist of a pure population of predominantly dispersed follicular cells with oncocytic features (Hurthle cells). Individual cells have enlarged round nuclei, fine chromatin, prominent nucleoli, and abundant finely granular eosinophilic cytoplasm. Nuclei are centrally to eccentrically placed, and some cells are binucleated. Occasional scattered Hurthle cells have markedly enlarged nuclei greater than twice the diameter of others. Some crowded clusters of Hurthle cells are seen as well as focal groups with transgressing blood vessels. The background contains blood and scant naked nuclei. Focally, a crowded group of hyperchromatic spindle cells is present.


Figure 6A - Aspirate smears were cellular and composed of a pure population of follicular cells with oncocytic features (Hurthle cells). Individual cells were large with enlarged, round nuclei, fine chromatin and prominent nucleoli. (H&E stain, high power)

Figure 6B - Occasional scattered Hurthle cells have markedly enlarged nuclei, greater than twice the diameter of others. (H&E stain, high power)

Figure 6C - Crowded clusters of Hurthle cells are seen as well as individual cells. Focally a crowded group of hyperchromatic spindle cells is present. (H&E, medium power)

Histology and Clinical Follow-Up:
The patient had a total thyroidectomy which revealed a mass that completely replaced the right thyroid lobe with extension into extrathyroidal soft tissues. Histologic examination revealed a widely invasive Hurthle cell carcinoma, and focally (less than 5% of the tumor), a markedly atypical spindle cell component representing anaplasic transformation was present. The patient died of local disease recurrence approximately 2 months after surgery.

Discussion: Hurthle cell carcinoma with focal anaplastic carcinoma
This case illustrates two important issues in thyroid aspiration cytology: 1) The limits of FNA as a screening test for Hurthle cell carcinoma, and 2) the clinical importance of recognizing an anaplastic component.

Hurthle cell neoplasms are composed of follicular cells where over 75% have oncocytic or oxyphilic features. These neoplasms comprise approximately 3% of all thyroid cancers and about 20% of all follicular tumors. Based upon their clinical behavior and molecular marker studies, Hurthle cell carcinomas may merit placement into a separate category of follicular neoplasms, although this is controversial. Relative to non-Hurthle cell neoplasms, a greater percentage of Hurthle cell neoplasms are malignant (10-32%). Also, Hurthle cell carcinomas relative to standard follicular carcinomas show a higher tendency to metastasize to lymph nodes (21%), and in some studies appear to be more clinically aggressive with a 5-year mortality rate as high as 50-60%.

As with non-oncocytic follicular neoplasms of the thyroid, Hurthle cell adenomas and Hurthle cell carcinomas cannot be distinguished by FNA, since like other follicular lesions of the thyroid, the histologic diagnosis depends upon recognizing the presence or absence of capsular and/or vascular invasion. Therefore both of these neoplasms are diagnosed by FNA as "suspicious for an oncocytic (Hurthle cell) neoplasm," and surgical excision is indicated. FNA correctly identifies the vast majority of Hurthle cell neoplasms as neoplastic. In addition, FNA distinguishes between true Hurthle cell neoplasms and other benign conditions with Hurthle cell changes such as Hashimoto's thyroiditis or multinodular goiter which can be accurately diagnosed as "benign" and for which surgical excision is not indicated. In general, cells from non-neoplastic Hurthle cell nodules form orderly flat sheets, and they often lack extensive distinct nucleoli. The typical cytologic features of true Hurthle cell neoplasms include a cellular smear with a uniform population of dyshesive cells with abundant granular cytoplasm, distinct red nucleoli, and finely granular chromatin. Nuclei are often eccentrically placed giving a plasmacytoid appearance, and binucleation is common. Colloid tends to be scant, and chronic inflammation is absent or sparse. The cells can appear bland, and often form crowded three-dimensional aggregates as well as occasional sheets with well-defined cell borders and follicles; transgressing vessels may be present within groups of cells. Important entities to exclude from the differential diagnosis of Hurthle cell neoplasms include: medullary carcinoma, oncocytic variant of papillary carcinoma, metastatic renal cell carcinoma, and parathyroid adenoma.

Unfortunately, as with non-oncocytic follicular lesions, FNA as a screening test for Hurthle cell carcinoma is markedly limited in its application by the fact that over 70-90% of lesions diagnosed as "suspicious" are in fact benign when surgically resected - A sensitive and specific ancillary marker for Hurthle cell carcinomas has yet to be identified. Recently, Dr. Renshaw identified a "new and improved" set of cytologic criteria for diagnosing Hurthle cell neoplasms. He suggests that by using a more specific set of 5 criteria, an improved specificity without loss of sensitivity can be obtained in the cytologic diagnosis of Hurthle cell neoplasms. These 5 criteria include: predominantly Hurthle cells and scant colloid, and at least one of either small cell dysplasia, large cell dysplasia, crowding, and dyshesion.

With regard to the anaplastic component present in the current case, the cytologic diagnosis of anaplastic carcinoma has extremely critical implications for the clinical management of a thyroid nodule since it may dictate whether or not surgical intervention will be used. The rare cases of anaplastic carcinoma associated with longer term survival (and where surgery may be considered as a therapeutic option) are usually those where the anaplastic carcinoma is small or forms a minor component of a well differentiated tumor.

Anaplastic carcinoma is among the most aggressive human malignancies with greater than 90% mortality and a dismal median survival of approximately 4-7 months. This carcinoma can arise de novo but in up to 80% of cases is associated with dedifferentiation of a precursor thyroid lesion such as in this case where it arose from de-differentiation of the Hurthle cell carcinoma. Anaplastic carcinoma represents approximately 2-5% of all thyroid carcinomas, and typically occurs in older patients (mean age 66 years), more commonly in women. At presentation, most tumors are large, and widely invasive with extension into extrathyroidal soft tissue, tracheal involvement, regional spread to lymph nodes, and sometimes distant metastasis. More than 50% of cases are inoperable at presentation.

Given adequate clinical and radiologic information, FNA of anaplastic carcinoma is highly accurate owing to its obvious malignant cytologic features. The cytologic appearance includes a cellular smear with malignant-appearing, sometimes bizarre, spindled and multinucleated tumor giant cells in groups and as single cells in a background of tumor diathesis (blood, fibrin, acute inflammation, necrotic cells, and debris). The nuclei of undifferentiated thyroid carcinoma are highly pleomorphic with dark, irregular chromatin clumping, macronucleoli, and occasional intranuclear pseudoinclusions. Numerous mitoses and abnormal mitotic figures may be seen.

At times the diagnosis of anaplastic carcinoma can be challenging since in some cases, only portions of the aspirate will show the anaplastic cyto-phenotype, while other portions of the cytologic specimen will consist of conventional well- or poorly differentiated carcinoma as in the case presented here. Clearly, thorough sampling of clinically suspicious lesions is important to avoid missing an anaplastic component. Fortunately, these are the rare cases where surgical intervention may be warranted to attempt a cure. When focal, anaplastic carcinoma must be distinguished from other spindled and atypical cells especially reactive elements. Immunocytochemical stains are helpful when adequate material is available for study. Most cases will exhibit keratin positivity helping to exclude a reactive fibroblastic, histiocytic, or vascular component of an otherwise well-differentiated thyroid neoplasm. In addition, anaplastic carcinomas will usually lack thyroglobulin and TTF-1 immunoreactivity that are seen in an adjacent well- or poorly-differentiated follicular carcinoma, and they will often exhibit overexpression of P53.

References

  1. Brooke PK, Hameed M, Zakowski MF. Fine-needle aspiration of anaplastic thyroid carcinoma with varied cytologic and histologic patterns: a case report. Diagn Cytopathol 1994;11:60-3.
  2. Gonzalez JL, Wang HH, Ducatman BS. Fine-needle aspiration of Hurthle cell lesions. A cytomorphologic approach to diagnosis. Am J Clin Pathol 1993;100:231-5.
  3. Guarda LA, Peterson CE, Hall W, Baskin HJ. Anaplastic thyroid carcinoma: cytomorphology and clinical implications of fine-needle aspiration. Diagn Cytopathol 1991;7:63-7.
  4. Kini SR, Miller JM, Hamburger JI. Cytopathology of Hurthle cell lesions of the thyroid gland by fine needle aspiration. Acta Cytol 1981;25:647-52.
  5. Nguyen G, Husain M, Akin MM. Cytodiagnosis of benign and malignant Hurthle cell lesions of the thyroid by fine-needle aspiration biopsy. Diagn Cytopathol 1999;20:261-65.
  6. Renshaw AA. Hurthle cell carcinoma is a better gold standard than Hurthle cell neoplasm for fine-needle aspiration of the thyroid. Cancer Cytopathol. 2002;96:261-6.
  7. Vodanovic S, Crepinko I, Smoje J. Morphologic diagnosis of Hurthle cell tumors of the thyroid gland. Acta Cytol 1993;37:317-22.
  8. Yang YJ, Khurana KK. Diagnostic utility of intracytoplasmic lumen and transgressing vessels in evaluation of Hurthle cell lesions by fine-needle aspiration. Arch Pathol Lab Med 2001;125:1031-35.