Currently there are only few reported cases of metastases to the axillary lymph nodes in patients with thyroid cancer and, to our knowledge, this is the first report on metastasis to the axillary lymph nodes from follicular thyroid carcinoma. In the previously published cases, the primary tumours were mostly papillary thyroid carcinomas [2
], and less frequently medullary thyroid carcinoma [7
], sclerosing mucoepidermoid carcinoma with eosinophilia [10
], mucin-producing poorly differentiated adenocarcinoma [11
] and muco-epidermoid carcinoma [12
]. In most of the previously reported cases, axillary lymph node metastases occurred in the context of disseminated disease, almost always as late recurrence. In our case it occurred at the presentation in the context of a disseminated follicular thyroid carcinoma. Some authors hypothesized that retrograde spread of the tumor into the axilla could be related to the blockage of the cervical lymphatic vessels by metastatic lymph nodes or by surgical manipulation [5
]. This hypothesis seems consistent with our case. In our patient, in fact, there was a massive lymph nodes involvement of the neck levels IV and V. The axillary fat is contiguous with that of the inferolateral neck, so we hypothesize that the spread to the axillary lymph nodes could be related to the obstruction of lymphatic vessels. Follicular thyroid cancer has a high propensity to invade blood vessels and show-up as distant metastases, but the role of angiogenesis and lymphangiogenesis in thyroid cancer pathogenesis has not been clarified [13
]. So, in the present case, it can't be established whether the axillary lymph node involvement is related to an hematogenous dissemination.
The pathological features of the neoplasm represent the second interesting aspect of our case. Primary mucinous carcinoma of the thyroid is extremely rare [14
]. In the WHO classification, mucinous carcinoma of the thyroid is defined as a tumor characterized by clusters of neoplastic cells surrounded by extensive extracellular mucin deposit. On the other hand, many thyroid neoplasms, such as follicular adenoma, may produce mostly intracellular mucins, in particular the "signet ring cell" variant; moreover mucus may be produced also by carcinomas of papillary, follicular, medullary and mucoepidermoid subtypes. Mlynek identified mucin in 50% and 35% of papillary and follicular tumors, respectively, and emphasized that mucin production by follicular-derived tumors can be commoner than usually thought [15
]. In the current case, "signet ring cells" and mucin-producing areas were frequently found.
To date, only four cases have been reported of follicular-derived thyroid carcinomas which show signet ring features: three by Schröder et al. [16
] and one by Fellegara et al. in a poorly differentiated Hürthle cell carcinoma [17
]. In a previous report, Gherardi et al. has presented convincing evidence that "signet ring cells" contained altered or abnormal thyroglobulin [18
]. Fellegara and Rosai [17
] have reported how vacuoles result from intracytoplasmic accumulation of thyroglobulin as demonstrated by immunohistochemistry, while they appear bordered by microvilli on electronic microscopy. This may suggest that they are intracytoplasmic follicular lumina rather than secretion vacuoles and represent the morphologic expression of an arrest in folliculogenesis [17
In the present case, both thyroglobulin and mucoid secretion were demonstrated in the "signet-ring cells", according to previous reports [16
]. Interestingly, microscopic examination of axillary metastases showed neoplastic cells with ample granular eosinophilic cytoplasm and pleomorphic nuclei, the typical features of Hürthle cell carcinomas, similarly to two previously reported cases of "signet ring cells" carcinoma [16
]; on the other hand, the "signet ring cell" features, in our case, were prominent in the primary thyroid tumor, while they were lacking in metastases. "Signet-ring cells" are hence to be considered a morphologic feature shared by both benign and, more rarely, malignant follicular thyroid neoplasm, and probably reflect an arrest in folliculogenesis. Our case is one of the rare "signet ring cells" carcinomas so far described.