—  SPECIALTY CONFERENCE  —

Surgical Pathology

Case 5 - Thyroid, Thyroidectomy - Medullary Carcinoma, Oncocytic Pigmented Variant

Juan Rosai
Centro Diagnostico Italiano, Milano, Italy
and Genzyme, Inc., New York, NY


Click on each slide thumbnail image for an enlarged view
Case History:
52-year-old male with bilateral thyroid nodules, which had been demonstrated on an MRI done for work-up of shoulder pain. The larger nodule was located in the left lobe; it measured 4.2 x 3.7 x 4.9 cm, and according to the patient it had been present for about 10 years. A FNA of this nodule was interpreted as "hypercellular, with changes suggestive of a neoplasm". A total thyroidectomy was carried out.

Grossly, the left thyroid lobe contained an encapsulated lobulated, solid, pink-tan mass measuring 4.5 cm in greatest diameter. The right lobe had a multinodular appearance with focal cystic changes.

The Seminar sections are from the nodule located in the left lobe.


Case 5 - Figure 1 - Peripheral portion of the thyroid tumor showing evidence of capsular invasion.
Case 5 - Figure 2 - Medium-power view of tumor showing solid nests and colloid-containing follicles.
Case 5 - Figure 3 - High-power view showing large size of tumor cells, abundant granular amphophilic cytoplasm and centrally located nucleus. The nucleoli are not particularly prominent. A thick fibrous band is seen in one corner of the photograph.


Diagnosis:
Thyroid, Thyroidectomy - Medullary Carcinoma, Oncocytic Pigmented Variant

Discussion:
This thyroid tumor exhibits a predominantly solid pattern of growth associated with scattered follicular formations. It is composed of large cells with an oncocytic appearance and it features both capsular and blood vessel invasion. In all likelihood, it would have diagnosed as Hurthle cell carcinoma (oncocytic follicular carcinoma) were it not for the fact that the pathologist who first examined the case noticed the focal presence of a brown pigment consistent with melanin in the cytoplasm of some tumor cells. The fact that the pigment in question was melanin rather than hemosiderin was confirmed with the Fontana-Masson stain (positive) and the Perls' iron stain (negative). This led to a critical reevaluation of the case, the submission to a consultant, the performance of a battery of immunostains, and the realization that this thyroid oncocytic tumor was not composed of follicular cells but of C cells, and that it was therefore a oncocytic and pigmented variant of medullary carcinoma. Specifically, the tumor cells proved to be strongly immunoreactive to calcitonin, chromogranin, synaptophysin, CEA, TTF-1 and mitochondrial antigen, while negative for thyroglobulin.

Oncocytic medullary carcinoma is one of the great simulators in thyroid pathology. I have been caught more than once, in the sense of having made a diagnosis of Hurthle cell carcinoma only to be told later by the referring pathologist that the serum calcitonin levels were sky-high and being later confronted with an embarrassingly strong positivity for calcitonin. At one point, I even considered the possibility of requesting a calcitonin stain in all oncocytic thyroid tumors. Since this would be overkill, I opted for the more rational approach of doing the stain selectively in the oncocytic tumors in which the H&E appearance offers some clue as to the possible C cell nature of the tumor. These clues are:
  1. The presence of sharply outlined fibrohyaline bands in between the tumor cells, resulting in the formation of incompletely circumscribed tumor nests;

  2. The fact that the cytoplasmic granularity does not have the bright eosinophilic quality of that typically seen in oncocytic follicular cells but rather an amphophilic (and focally basophilic) character. This is due to the presence in the cytoplasm of the tumor cells of two contrasting colors: the red color provided by the mitochondriae, and the blue color derived from the neurosecretory granules. It is the relative amounts of these two components that determines the color tone one sees in the sections;

  3. The presence of non-oncocytic areas, in which the tumor acquires an appearance more typical of medullary carcinoma, including sometimes the deposition of amyloid;

  4. The presence, as in the Seminar case, of a neuroepithelial-type product such as melanin. Alas, this is a very rare occurrence.
The oncocytic type is only one of the many morphologic or phenotypic variations that medullary carcinoma of the thyroid can exhibit. Others include: carcinoid-like, paraganglioma-like, trabecular, glandular (tubular and follicular), and pseudopapillary. Other more or less distinct variants are the inflammatory type, the anaplastic or giant cell type, the true papillary type, the mucinous or amphicrine type, the clear cell type, the squamous variant, the small cell type (which can resemble either a pulmonary small cell carcinoma or a neuroblastoma), and the already mentioned pigmented (melanin-producing) variant [1, 2, 3, 4, 5, 6].

It should also be remembered that medullary carcinoma can be seen in combination with follicular carcinoma (mixed medullary-follicular carcinoma) and –less commonly- with papillary carcinoma (mixed medullary-papillary carcinoma) [10, 11, 13]. Finally, it should be realized that not all neuroendocrine tumors involving the thyroid gland are of necessity medullary carcinomas. There are primary thyroid paragangliomas [7], primary high-grade neuroendocrine carcinomas not otherwise specified (calcitonin-negative), and neuroendocrine carcinomas of various sites and degree of differentiation metastatic to the thyroid (such as pulmonary carcinoid tumors) [9].

In the presence of a medullary thyroid carcinoma of any type, a molecular search for a germ cell line mutation of the RET oncogene needs to be carried out in order to detect familial cases of the disease (MEN1, MEN2, or familial medullary carcinoma) [8].

In regard to the production of melanin by the tumor, it is of interest that this property is not exhibited in equal fashion by all types of neuroendocrine tumors but primarily by those having a more "neural" phenotype, which for some as yet unexplained reason are usually located in the anatomic region that – besides the thyroid gland – includes larynx, lung and thymus [10].

A final comment relates to the possible sequence of events in the development of oncocytic medullary carcinoma. For the follicular neoplasms with oncocytic features, Sobrinho-Simoes and coworkers have proposed two theoretical pathways [11]
  1. First the normal follicular cell becomes oncocytic and in a second step it becomes neoplastic. The result would be a true oncocytoma;

  2. First the normal follicular cell produces a neoplasm of one type or another (papillary carcinoma, follicular carcinoma, etc.) and in a second step the neoplasm acquires oncocytic features. In such instance, the tumor should be viewed as a specific type of thyroid neoplasm (let's say papillary carcinoma) with secondary oncocytic features. Assuming that this hypothesis is accurate and that it could be applied to tumors of C cells, it would seen that the second scenario would apply, i.e., normal C cell → medullary carcinoma → oncocytic medullary carcinoma (medullary carcinoma with oncocytic features).

References

  1. Ben Romdhane K, Khattech R, Ben Othman M, Gamoudi A, Ammar A, Cammoun M: Melanin production in medullary thyroid carcinoma. Histopathology 27: 569-572, 1995.

  2. Beerman H, Rigaud C, Bogomoletz WV, Hollander H, Velhuizen RW: Melanin production in black medullary thyroid carcinoma. Histopathology 16: 227-234, 1990.

  3. Dominguez-Malagon H, Macias-Martinez V, Molina-Cardenas H, Suster S: Amphicrine medullary carcinoma of the thyroid with luminal differentiation: Report of an immunohistochemical and ultrastructural study. Ultrastruct Pathol 21: 569-574, 1997.

  4. Dominguez- Malangon H, Delgado-Chavez R, Torres-Najera M, Gould E, Albores-Saavedra J: Oxyphil and squamous variants of medullarythyroid carcinoma. Cancer 63: 1183-1188, 1989.

  5. Harach H, Bergholm U: Medullary (C cell) carcinoma of the thyroid with features of follicular oxyphilic cell tumors. Histopathology 21: 378-379, 1992.

  6. Harach H, Bergholm U: Small cell variant of medullary carcinoma of the thyroid with neuroblastoma-like features. Histopathology 21: 378-379, 1992.

  7. La Guette J, Matias-Guiu X, Rosai J: Thyroid paraganglioma: A clinicopathologic and immunohistochemical study of three cases. Am J Surg Pathol 21: 748-753, 1997.

  8. Matias-Guiu X: RET protooncogene analysis in the diagnosis of medullary thyroid carcinoma and multiple endocrine neoplasia type II. Adv Anat Pathol 5: 196-201, 1998.

  9. Matias-GuiuX, LaGuette J, Puras-Gil AM, Rosai J: Metastatic neuroendocrine tumors to the thyroid gland mimicking medullary carcinoma. A pathologic and immunohistochemical study of six cases. Am J Surg Pathol 21: 754-762, 1997.

  10. Matias-Guiu X, Caixas A, Costa I, Cabezas R, Prat J: Compound medullary-papillary carcinoma of the thyroid: true mixed tumor. Histopathology 25: 183-184, 1994.

  11. Papotti M, Negro F, Carney JA, Bussolati G, Lloyd RV: Mixed medullary-follicular carcinoma of the thyroid. A morphological, immunohistochemica and in situ hybridization analysis of 11 cases. Virchows Arch 430: 397-406, 1997.

  12. Rosai J: An evolutionary view of neuroendocrine cells and their tumors. Int J Surg Pathol 9: 87-92, 2001.

  13. Sobrinho-Simoes M: Mixed medullary and follicular carcinoma of the thyroid. Histopathology 23: 287-290, 1993.

  14. Sobrinho-Simoes M, Maximo V, Vieira De Castro I, Fonseca E, Soares P, Garcia-Rostan G, Cardoso de Oliveira M: Hurthle (oncocytic) cell tumors of thyroid. Etiopathogenesis, diagnosis and clinical significance. Int J Surg Pathol 13: 29-36, 2005.