University of Texas M.D. Anderson Cancer Center Houston, Texas 77030
Address correspondence and requests for reprints to: Robert F. Gagel, M.D., Section of Endocrine Neoplasia and Hormonal Disorders, Box 15, 1515 Holcombe Boulevard, Houston, Texas 77030. E-mail: bob_gagel{at}isqm.mda.uth.tmc.edu
![]() |
Introduction |
---|
![]() ![]() ![]() |
---|
Several reports over the past 2 yr employing sensitive and specific calcitonin assays provide some answers. Reports from Pacini et al. (1), Rieu et al. (2), and a study in this issue of JCEM (3) (see page 338) have evaluated the significance of basal serum calcitonin measurements in over 3,000 patients with nodular thyroid disease. The first 2 reports are similar in that patients with elevated basal calcitonin levels or abnormal FNA results underwent surgical exploration of the thyroid gland. In the study by Pacini et al. (1), 8 of 1,385 patients had elevated basal calcitonin levels ranging from 5510,000 pg/mL. In the study by Rieu et al. (2), 4 of 469 patients had elevated basal calcitonin levels ranging from 702,838 pg/mL. All patients in these 2 studies with an elevated basal calcitonin, as defined in the individual assay, were found to have medullary thyroid carcinoma (MTC). Equally important, only 3 of 12 patients identified with MTC by calcitonin measurements would have been identified by fine needle aspiration.
What these first 2 studies do not address is the relevance of a minimal elevation of the serum calcitonin. Seven of 12 patients in these 2 studies had basal serum calcitonin values greater than 200 pg/mL, a level highly suspicious for MTC. In addition, the design of these 2 studies, in which only patients with abnormal calcitonin values or abnormal FNA cytology underwent thyroid exploration, makes it difficult to determine the true prevalence of MTC.
The current report from Niccoli et al. (3) provides a more comprehensive assessment of this issue. A total of 1,167 patients were referred for surgical treatment of nodular thyroid disease including multinodular goiter, multinodular toxic goiter, and single "euthyroid" or "autonomously functioning" nodules. Each of the patients had a basal serum calcitonin measured by a specific and sensitive two-site immunoradiometric assay for human calcitonin (4), eliminating the low frequency problem of artifactual elevations of the serum calcitonin observed with single-site radioimmunoassays (5). Approximately 1/10 of the patients with normal basal serum calcitonin values, selected by unclear criteria, also had serum calcitonin measurements after pentagastrin stimulation. All 1,167 patients underwent surgical removal of the thyroid gland and histologic analysis by serial sectioning of the thyroid gland to detect small or microscopic abnormalities, an ambitious undertaking.
Several observations are worth emphasizing. Sixteen of 1,167 patients had medullary thyroid carcinoma, a prevalence of 1.37%, higher than that previously reported in the literature. It is difficult to determine if this is the true prevalence as the pathologists were apparently not blinded to the calcitonin results, and an even more thorough search was done by staining with an anticalcitonin polyclonal antibody to detect microscopic MTC in patients with abnormal basal or pentagastrin-stimulated calcitonin values. Therefore, 1.37% may actually be an underestimation of the true prevalence.
Thirty-four of 1,167 patients had an abnormal basal serum calcitonin value ranging from 1137,000 pg/mL. Fourteen of the 34 patients or 41% were found to have a focus of MTC. Significantly, in 9 of the 14 patients with MTC, the serum calcitonin values were below 200 pg/mL, ranging between 11 and 200 pg/mL. The size of the MTCs in these patients ranged between 0.125 and 1.0 cm. Postoperative evaluation by pentagastrin testing suggests that all patients with microscopic MTC had normal basal and pentagastrin-stimulated calcitonin values with 24 yr of followup.
Another interesting point, not commented upon by the authors, is that 22 of 34 patients, or 65%, with an elevated basal calcitonin had some type of thyroid carcinoma identified in the gland, either MTC or differentiated thyroid cancer. One does not think of an elevated serum calcitonin as an indicator of follicular cell neoplasms, but it is well known that C cell hyperplasia may occur adjacent to a follicular adenoma/carcinoma or papillary thyroid carcinoma (6, 7).
Another finding is the coexistence of thyroiditis and MTC in 50%, 8 of 16 cases with elevated basal CT values and 2 of 7 patients with abnormal pentagastrin-stimulated CT values. As the authors point out, there is a known association between Hashimotos thyroiditis, C cell hyperplasia, and increased plasma or serum CT values (8, 9), suggesting the possibility of a cause (thyroiditis) and effect (C cell hyperplasia and MTC) relationship between thyroiditis and malignancy (10, 11). The data set is striking in that it appears that an elevated basal serum calcitonin level, regardless of the degree of elevation, is associated with an increased risk of thyroid cancer.
The significance of a small medullary thyroid carcinoma in this context is less clear. In hereditary MTC, abundant evidence points to the continued growth and metastasis of an early C cell lesion (12, 13, 14). No such evidence exists for small sporadic medullary thyroid carcinomas. Could these lesions be analogous to the incidental papillary thyroid carcinoma identified at autopsy and associated with a low incidence of morbidity and mortality (15)? The infrequency of detection of small medullary thyroid carcinomas suggests we will never obtain a satisfactory answer to this question. However, the lack of effective therapy for medullary thyroid carcinoma other than surgery suggests that it is appropriate to remove a small medullary thyroid carcinoma.
Another point not commented upon by the authors is the cost of basal calcitonin testing, an issue of increasing concern in the United States. The commercial charge for a single calcitonin measurement in the USA averages $100. Therefore, to detect the 16 cases of MTC in the report by Niccoli et al. (3), it would be necessary to spend at least $100,000 or $5,800 per diagnosed case. If we consider that 5 of these cases were detected by FNA, the cost rises to $12,500 per diagnosed case. Although this price may seem high, identification of MTC as a palpable thyroid nodule is associated with a high percentage of metastasis, frequently requiring additional treatment. The costs of initial basal calcitonin screening should be balanced against those of multiple surgical procedures, radiation, and chemotherapy frequently used in a patient with metastatic disease.
Another important component in any strategy to utilize calcitonin measurements for decision making is the specificity of the assay. Earlier studies with a single-site radioimmunoassay have documented nonspecific increases of the serum calcitonin in normal subjects, and these artifactual elevations probably occur more commonly than reported (5). Niccoli et al. (3) have utilized a 2-site immunoradiometric assay for their calcitonin measurements (4), thereby eliminating this type of artifact and giving this study even greater impact. At the time of this writing, no commercial laboratory in the United States is using a 2-site assay, although at least two will be converting in the near future, and the kit used in these studies is available in the United States. A question that should be asked in each of these situations is whether there is another plausible source of calcitonin other than the thyroid, such as benign or malignant lung (16, 17, 18), breast (19), or other endocrine tumors (20).
Finally, should patients who have a nodular goiter and serum calcitonin elevation have molecular testing for c-ret proto-oncogene mutations causative for hereditary MTC (21, 22, 23)? Five to seven percent of patients with apparent "sporadic" medullary thryoid carcinoma have germ-line mutations of the c-ret proto-oncogene that affect either the extracellular cysteine-rich or the intracellular tyrosine kinase domains (24). This has led to the recommendation that c-ret proto-oncogene testing be performed routinely in all patients with sporadic MTC. In patients with proven medullary thyroid carcinoma this makes sense because the 57% detection rate is doubled or tripled if one includes previously undetected gene carriers within these families (24). Since only 41% of the patients with basal calcitonin elevations and multinodular goiters in the report by Niccoli et al. (3) have medullary thyroid carcinoma, no more than 1 or 2 patients out of every 100 screened would be expected to carry a germline mutation, a very low and probably not cost-effective yield.
In conclusion, a substantial experience suggests that measurement of basal calcitonin values in patients with nodular thyroid disease has merit. Whether these studies, performed in France and Italy, are applicable to the population in the United States or other countries is unclear. Nonetheless the finding of MTC in 41% and either MTC or differentiated thyroid carcinoma in 65% of patients with basal calcitonin elevations and multinodular goiter is impressive. It may be appropriate to include serum calcitonin measurements in the routine evaluation of a nodular thyroid gland and consider thyroidectomy in patients with nodular thyroid disease and an elevated basal serum calcitonin.
Received July 15, 1996.
Accepted November 26, 1996.
![]() |
References |
---|
![]() ![]() ![]() |
---|