University of Pennsylvania Medical Center Philadelphia, Pennsylvania 19104
Address correspondence and requests for reprints to: Virginia A. LiVolsi, MD, Department of Pathology and Laboratory Medicine, University of Pennsylvania Medical Center, 3400 Spruce Street, Philadelphia, Pennsylvania 19104.
![]() |
Introduction |
---|
![]() ![]() ![]() |
---|
Pathologically, C cell hyperplasia in the familial setting of an inherited autosomal dominant process is found most prominently in the upper two thirds of the thyroid gland in the lateral lobes. In this type of C cell hyperplasia, the C cells are noted as groups of clear large cells in a peri- or parafollicular location on hematoxylin and eosin staining. Sometimes the cytologic features of the C cells show atypia including nuclear atypia.
In familial associated C cell hyperplasia, the C cells may be found in other areas of the gland, although very rarely in the isthmus. In these conditions, the evidence of C cell hyperplasia correlates well with elevations of serum calcitonin either basal or after stimulation by pentagastrin or calcium.
The term C cell hyperplasia has been challenged by Carney et al. (1) who state that the lesion is really preinvasive carcinoma. They prefer the term C cell disease rather than C cell hyperplasia. Indeed, familial C cell hyperplasia is a preneoplastic lesion and constitutes carcinoma in situ of the thyroid gland parafollicular cells.
In broad pathologic terms, the concept of carcinoma in situ (CIS) represents the accumulation of various genetic events from a normal epithelium to a fully malignant one. These genetic events are manifested microscopically by increasing cytologic atypia, nuclear anaplasia, easily identified mitotic activity, abnormal mitotic figures, and on occasion, necrosis. Changes in vascularity, so called angiogenesis, also occur.
The CIS concept is best illustrated in uterine cervix squamous carcinoma and duct carcinoma of the breast. In each of these situations (as in other epithelia in the human body that can give rise to invasive cancers), the diagnosis of in situ carcinoma by established pathologic criteria is relatively straightforward. The clinical meaning and therapy have also been well defined. It is known from large and long-term follow-up epidemiological studies, as well as studies in which the diagnosis of CIS was not correctly identified initially (this is especially true in the breast), that a significant percentage of cases of CIS will progress through additional presumably genetic mutations to breach the light microscopy basement membrane and invade the underlying stroma of the organ of origin. Beginning as microinvasive carcinoma and ending as fully invasive carcinoma, this process continues with ultimate development of metastatic disease. It has been shown that even so called microinvasive carcinomas can on occasion metastasize to regional lymph nodes, although the smaller the focus of microinvasion, the less risk of this occurrence.
In the background of this familial setting, C cell hyperplasia is, in fact, C cell carcinoma in situ or C cell neoplasia. Gill et al. (2) describe early invasion (medullary microcarcinoma) in very young patients with familial C cell hyperplasia. Some of these early carcinomas are capable of metastasizing (2).
Komminoth et al. (3) identified differences between so called primary C cell hyperplasia associated with familial medullary carcinoma, by immunohistochemical staining for expression of polysialic acid (a post-translation modification of neural cell adhesion molecule). This is identified in primary C cell hyperplasia and in medullary thyroid carcinoma. It is not found immunohistochemically in other thyroid tumors, malignant or benign, or in so called secondary C cell hyperplasia (see below) or normal C cells. These authors suggest that this protein is a marker for medullary carcinoma and may be able to distinguish between primary and secondary C cell hyperplasia.
It is still debatable how easy it is for pathologists to identify the lower limit of medullary carcinoma, that is, microinvasive medullary thyroid carcinoma and the upper limit of C cell hyperplasia/C cell neoplasia. DeLellis and Wolfe (4) note that C cell hyperplasia ranges from a diffuse increase in the cells to nodules of C cells that eventually replace follicles so that they become devoid of colloid. They indicate that once the basement membrane of the follicle is breached, medullary carcinoma should be diagnosed; however, Carney et al. (1) note that it is not always obvious that the basement membrane has been crossed. In addition, it is known that C cell hyperplasia will show diffuse and intense immunereactivity for calcitonin, greater than that for medullary carcinoma (4). However, this is a matter of degree and also of technical expertise; in any individual case, the immunostaining intensity may not be helpful in distinguishing early invasion, that is medullary carcinoma, from florid C cell hyperplasia/C cell neoplasia.
McDermott et al. (5) indicate that immunostaining for basement membrane collagen type IV may help in this distinction; however, these results are preliminary and have not been widely used.
If this familial C cell hyperplasia or primary C cell hyperplasia truly represents carcinoma in situ of the endocrine system, it is a unique pathologic lesion. Carcinoma in situ of the endocrine system has not been defined with the above exception. Virtually all endocrine proliferations can be conveniently subdivided into hyperplasias, that is, physiologic or pathophysiologic response to a driving force, or neoplasias i.e. an autonomous clonal event. Hyperplasias in the thyroid can be a response to abnormal circulating immunoglobulins that affect and bind the TSH receptor complex (as in autoimmune hyperthyroidism); less frequently, a response to excess TSH from the pituitary because of either pituitary dysfunction or neoplasia; or they can be a secondary response of the thyroid to elevated pituitary TSH because of depressed production of thyroid hormone by the thyroid gland (inborn error of thyroid metabolism).
Nodules in the endocrine system therefore can be hyperplastic or neoplastic. Apel et al. (6) have shown that 70% of adenomatous or hyperplastic nodules in multinodular goiter are, in fact, clonal proliferations. These are however, all benign based on the evaluation of large numbers of patients who have these nodules and who do not develop malignant clinical courses.
Proliferations that are encapsulated and are correctly called adenomas are virtually always clonal proliferations in the thyroid when they have been studied (7). When they are diagnosed and removed, they apparently always behave in a benign fashion if they do not have evidence of invasion histologically. This is true also of encapsulated neoplasms of adrenal, parathyroid, and pituitary. Anecdotal reports (often discussed at pathology slide seminars!) have been described as apparent follicular adenomas of the thyroid, in which follicular carcinomas arise. It is indicated in these cases that there was a preexisting follicular adenoma (often the clinical history is appropriate to that fact), and there has been recent growth. When one examines such a lesion under the microscope, one can see an area of follicular patterned tumor that has invaded into its capsule or into surrounding vessels. Similarly, rare instances of parathyroid carcinoma apparently arising in the background setting of parathyroid adenoma are known. If these transitions (transformations) can occur in clonal lesions, evaluation of C cell hyperplasia/neoplasia in thyroids from MEN-2 patients for clonality would confirm these lesions as neoplasms.
An additional piece of evidence that there may be a transition from benign to malignant lesions, that is, a carcinoma in situ in the endocrine system, comes from data on flow cytometric analyses of various endocrine proliferations. In the thyroid, Joensuu et al. (8) have shown that up to 25% of histologically and clinically benign follicular adenomas have aneuploid cell populations on flow cytometry. However, these lesions behave clinically in a benign fashion and histologically do not show any evidence of invasive characteristics. Whether or not these lesions represent carcinoma in situ in the endocrine system has not been decided; however, this technique may not be applicable to all cases. On the basis of light microscopic examination, one cannot define which of these lesions will show aneuploidy and which will show normal DNA content. It is also critical to note that the great majority of differentiated thyroid carcinomas are diploid by flow cytometric analysis for DNA.
The second area that is problematic in the pathologic definition of C cell hyperplasia is much more difficult. This is, what is the upper limit of normal cell mass and/or number and what is the lower limit of C cell hyperplasia? Outside of the setting of familial medullary carcinoma and its variants (MEN), C cell hyperplasia (so called secondary C cell hyperplasia) (2) has been described in aging individuals, in hyperparathyroidism, hypergastrinemia, and near follicular derived tumors, usually malignancies (9, 10). Historically, normal C cells were described as occurring singly or in small groups of 35 cells; in an adult thyroid gland, C cells numbering greater than 10 per low power field were labeled C cell hyperplasia. Numerous studies over the years, predominantly the elegant work of DeLellis and Wolfe and their colleagues (9) have indicated that one can see up to 50 C cells per low power field in areas of high C cell concentration in the adult human thyroid and that this may be unassociated with any other pathologic abnormality, any systemic disorder, or any evidence of hypercalcitoninemia. The work of Guyetant et al. (11) in thyroiditis indicated that greater than 40 C cells per square centimeter, or more than 50 C cells per 3 low power fields, define C cell hyperplasia, outside the setting of familial medullary carcinoma. However, these studies do not show correlation with serum calcitonin levels and are strictly pathologic definitions even though elegantly measured by morphometric techniques.
It is known by individual case reports (12) that one can have hypercalcitoninemia and C cell hyperplasia (very similar to that seen in C cell hyperplasia in familial medullary carcinoma, although as yet no one has studied this with the polysialic acid immunostain) in severe chronic lymphocytic thyroiditis. It is at least theoretically possible that such a phenomenon may occur as a result of depressed thyroid hormone status and elevated TSH that may be stimulatory to the C cells; this is, in fact, a true secondary C cell hyperplasia.
In the current issue of this journal, Guyetant et al. (see page 42) (13) present an elegant morphometric study of C cells in normal thyroids. They conclude that 33% of adult subjects (15% of women and 41% of men) fulfill the criteria of C cell hyperplasia, that is at least three x100 magnification fields containing more than 50 C cells. This was not correlated with elevation of serum calcitonin levels. In fact, the thyroids evaluated were derived from endocrinologically normal individuals. Whether or not 33% of the human adult population harbors this disorder of C cell hyperplasia or whether definitions of C cell hyperplasias are incorrect requires further analysis of normal thyroids and correlation with serum calcitonin levels. In fact, Perry et al. (10) state that numbers of C cells are not pathologically important, but rather whether or not they are cytologically atypical.
It is interesting that in this study, Guyetant et al. find that only one third as many women as men had C cell hyperplasia. If in fact this observation is confirmed, the relationship of this finding to the much higher incidence of osteoporosis in women as compared with men, should be explored by further studies.
Another important difference between primary C cell hyperplasia/C cell neoplasia and the secondary form is the rarity of progression of the latter to invasive medullary carcinoma (14). As noted above (2), C cell hyperplasia in the familial setting is preinvasive, which can give rise to carcinoma even with metastasis in very young patients.
Hence, the pathologic definition and clinical correlation of secondary C cell hyperplasia remains unclear.
Received October 21, 1996.
Accepted October 23, 1996.
![]() |
References |
---|
![]() ![]() ![]() |
---|