750 Physiologic Versus Neoplastic C-Cell Hyperplasia of the Thyroid Separation of Distinct Histologic and Biologic Entities Arrie Perry, M.D.’ Kyle Molberg, M.D.’ Jorge Albores-Saavedra, M.D.’ ’ Department of Pathology, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas. Address for reprints: Jorge Albores-Saavedra, M.D., Division of Anatomic Pathology, Department of Pathology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas. TX 75235-9072. Received August 11, 1995; revision received October 19, 1995; accepted October 27, 1995. SI 1996 American Cancer Society BACKGROUND. Although hyperplasia of C-cells has been described in association with various pathologic and physiologic conditions, criteria for its diagnosis are poorly defined. Both neoplastic and physiologic C-cell proliferations have been lumped together under the umbrella designation of C-cell hyperplasia (CCH), creating considerable confusion among clinicians and pathologists. METHODS. In order to compare the morphologic and inimunohistochemical characteristics of the two major types of CCH, we examined thyroid sections of 17 patients with familial forms of C-cell hyperplasia and/or neoplasia and tissue sections of 19 thyroid glands known to have reactive or physiologic CCH (at least 50 C-cells per one low power field, 1 0 0 ~ )Hematoxylin . and eosin (H & E) stained sections and immunohistochemical stains for calcitonin were assessed in each case. RESULTS. Physiologic or reactive CCH was not recognized with certainty on H & E stains in any of the cases due to morphologic similarities between C-cells and adjacent follicular cells. Detection of this form of hyperplasia, which was predominantly diffuse, required calcitonin imniunostains and quantitative analysis. Conversely, nodular and diffuse neoplastic CCH was easily identified with conventional H & E stains at the periphery of 11/12 (92%)familial medullary thyroid carcinomas (MTC).I n the other five cases, neoplastic C-cell hyperplasia was the only pathologic finding on thyroidectomy performed for elevated serum calcitonin levels detected via provocative biochemical screening or identification of the mutated RET protooncogene by genetic analysis. The C-cells in this neoplastic form of CCH were large, mildly to moderately atypical, and confined within the basement membrane of thyroid follicles. Moreover, these cells were cytologically indistinguishable from those of invasive MTC cells. CONCLUSIONS. Physiologic and neoplastic CCH are biologically and morphologically distinct entities. The former cannot be recognized with certainty with conventional stains and requires itnmunohistochemistv and quantitative analysis for diagnosis. The latter consists of mildly to moderately atypical C-cells that can be identified with H & E stained sections. Consequently, the number of C-cells is of no importance for the diagnosis of neoplastic CCH which is considered to be the precursor (medullary carcinoma in situ) of invasive medullary carcinoma. Cancer 1996; 77:750-6. 0 1996 American Cancer Society. KEYWORDS physiologic C-cell hyperplasia, medullary carcinoma in situ, MEN Ila, MEN Ilb, familial medullary carcinoma, irnmunohistochernistry. S ince the first description of calcitonin-producing C-cells’ and their malignant counterpart, medullary thyroid carcinoma (MTC), a number of questions regarding the C-cell proliferations known as C-cell hyperplasia (CCH) remain unanswered. CCH has been used for at least two C-Cell Hyperplasia/Perry et al. different biologic processes. The first is a misnomer used to describe the precursor of familial MTC (FMTC) associated with multiple endocrine neoplasia (MEN) IIa, MEN IIb, and FMTC unassociated with other manifestations of MEN."4 The :second form of CCH is a physiologic or reactive C-cell proliferation which has been reported in neonates,' in the e1derly,'"Ih and in patients with hyperparathyroidi~rn."~'~ Hashimoto's thyroiditi~,"-~~ previous hemithyr~idectomy,'~follicular thyroid neoplasms 25.'h and non-Hodgkin's lymphomas of the thyroid." Although poorly understood, the pathogenesis of this form has been attributed to overstimulation by thyroid-stimulating hormone (TSH),hypercalcemia, and other factors. In contrast to the neoplastic form, physiologic or reactive CCH is not associated with MTC. Studies of physiologic and neoplastic CCH based on conventional stains are almost nonexistent. As a result, the information provided in many articles has led some to believe that immunohistochemistry (IHC) or electron microscopy are indispensable for the diagnosis of both types of C-cell proliferations. The lumping of these two biologically different forms of CCH under the same diagnostic heading has led to considerable confusion among clinicians and pathologists. This confusion is compounded by lack of uniform diagnostic criteria. In this study, we emphasize the morphologic differences between physiologic and neoplastic CCH and discuss diagnostic criteria for each. MATERIALS AND METHODS Seventeen thyroidectomy cases with clinical or histologic evidence of familial C-cell neoplastic disease (MEN IIa, MEN IIb, and FMTC) were retrieved for further study (Group I) from the surgical pathology files at the University OF Texas Southwestern Medical Center, Dallas, and from 1he personal consultation files of one of the authors (JAS). Criteria lor establishing the familial form of C-cell neoplasia included appropriate family history, extrathyroid manifestations of MEN 11, and multifocal MTC or neoplastic CCH. In three patients, mutations of the RET proto-oncogene were identified in the peripheral blood. Neoplastic CCH was defined as the presence of large, mildly to moderately atypical, intrafollicular cells cytologically resembling those of MTC with immunoreactivity for calcitonin. Group I1 consisted of 19 examples of physiologic CCH including 17 previously reported cases associated with follicular neoplasms, and 2 associated with Hashimoto's thyroiditis. Physiologic CCH was defined according to criteria set by Albores-Saavedra et al.z5which require 50 or more cytologically bland intrafollicular calcitonin positive cells in at least one low power field (1OOX). All available clinical data and hematoxylin and eosin TABLE 751 1 Patient Population No. of cases Hypercalcitoninemiaa 17 10 8 6 3 6 4 NP Group 11 (Physiologic CCH) 19 3 Papillary carcinoma Follicular carcinoma Follicular adenoma Hashimoto's thyroiditis 10 2 5 2 Associated diagnosis ~~~ Group I (Neoplastic CCH) MEN Ila FMTC Histologic evidence of FMTC 1 NP 1 1 Basal or following pentagastrin stimulation. N P not Derformed; FMTC familial medullary thyroid carcinoma (H & E) slides were reviewed by one of the authors (JAS). Immunostains were performed in all cases using the standard avidin-biotin peroxidase method. The calcitonin antibody (dilution 1:100) was obtained commercially (Lipshaw, Pittsburg, PA). Appropriate positive MTC and negative controls were used. RESULTS Table 1 categorizes the patient populations in Groups I and I1 according to their associated diagnoses. Patients in Group I (neoplastic CCH) consisted of 11 males and 6 females who ranged in age from 3 to 61 years (mean age, 21 years). Patients in Group I1 (physiologic CCH) consisted of 7 males and 12 females who ranged in age from 17 to 54 years (mean age, 33 years). Eight patients in Group I had a family history of MEN IIa and six had a family history of FMTC. The remaining three patients had no family history of thyroid neoplasms, however, they did have histologic evidence of the genetic form of MTC. All three patients had multifocal MTC or neoplastic CCH and one patient had parathyroid hyperplasia in addition. None of the patients in Group I1 had either personal or family histories of MTC. Of the 17 patients in Group I, 10 had thyroidectomies for repeatedly elevated serum calcitonin levels with or without pentagastrin stimulation. Seven of these specimens had small non-palpable, bilateral medullary carcinomas varing from 0.1 cm to 0.7 cm in diameter, while three showed only bilateral CCH. Four of the seventeen patients had thyroidectomies for palpable masses. Three of these specimens revealed palpable medullary carcinomas ranging from 1.0 cm to 3 cm in diameter. The fourth patient had a 0.4 cm medullary carcinoma discovered incidentally adjacent to a palpable follicular adenoma. 752 CANCER February 15,1996 / Volume 77 / Number 4 power view Of neoplastic CCH showing diffuse and nodular growth patterns The C-celk are morphologically distinct from the adjacent follicular cells (xl25). FIGURE 2. Diffuse neoplastic CCH. The follicular cells (arrow) are being pushed upward and compressed by the larger neoplastic C-cells ( ~ 2 0 0 ) . FIGURE 3. Nodular neoplastic CCH. The follicles have been obliterated by neoplastic C-cells. Some of which show mild nuclear atypia ( ~ 2 0 0 ) . Three of the seventeen patients had thyroidectomies after the discovery of the mutated KET proto-oncogene in their blood. Two of these specimens had only CCH, while the third contained a 0.2 cm medullary carcinoma. Thyroidectomy in all Group I1 patients was for palpable masses although three of these patients also had elevated serum calcitonin. One was a female, age 19 years, with a papillary carcinoma. The second was a female, age 27 years, with Hashimoto's thyroiditis, hypothyroidism, and elevated serum TSH, and the third was a man, age 37 years, with a follicular adenoma. Neoplastic CCH was characterized by clusters of large, cytologically atypical round, polygonal, or spindleshaped, intrafollicular cells (Figs. 1-4). When present in association with MTC, the cells were cytologically identical to the invasive component. These cells were recognizable on H & E sections and were associated with strong FIGURE 4. Nodular and diffuse neoplastic CCH with strong calcitonin positivity ( ~ 2 0 0 ) . cytoplasmic staining for calcitonin (Fig. 4). Patterns of involvement included focal with partial replacement of the follicle, diffuse with complete replacement of the periphery of the follicle resulting in ring-like structures, or nodular with complete replacement and filling of the follicular lumen. The diffuse and nodular patterns were most common. Physiologic CCH consisted of an increased number of normal C-cells which resembled follicular cells and were, therefore, not recognizable with certainty on routine H & E stains. In five cases, there were occasional C-Cell Hyperplasia/Perry et al. FIGURE 5. Physiologic CCH in a case of Hashimoto's thyroiditis. A few scattered cells with abundant clear cytoplasm suggestive of C-cells are present ( ~ 1 2 5 ) . 753 FIGURE 6. Higher magnification of Figure 5 showing the intrafollicular large cells with clear cytoplasm suggestive of C-cells. Note that the nuclear features of these cells are similar to those of adjacent follicular cell ( ~ 2 0 0 ) . DISCUSSION FIGURE 7. Calcitonin immunostaining of physiologic CCH reveals many more C-cells than suspected from the H 8, E stain (x125). large intrafollicular cells with abundant granular or clear cytopl,ism suggestive of C-cells (Figs. 5 and 6). However, the nuclei in these cells were similar to those of adjacent follicular cells. These rare large cells identified o n H & E stain did not reflect the large number of calcitonin positive C-cells seen with IHC (Fig. 7). These calcitonin immunostains revealed increased numbers of C-cells (see diagnostic criteria in Methods), most commonly in a diffuse pattern. A nodular growth pattern was also seen in 2 of 20 examples of physiologic CCH. Physiologic CCH was usually localized to one region of the thyroid and was bilateral in only one case which was associated with Hashimoto's thyroiditis. In most cases associated with follicular neoplasms, the CCH was found in the non-neoplastic thyroid tissue adjacent to the neoplasms. Since the normal human C-cell was identified nearly 30 years ago,' there has been a lack of uniformity in criteria for defining CCH. Criteria have varied from subjective increases in C-cells as suggested by Wolfe et al.,"' to the presence of C-cell clusters or nodule^,^^^^^^" to quantitative assessments including a minimum of 50 C-cells per 50 low power fields (lOOx)," 4 C-cells per low power ) frequent grouping,'4 6 C-cells per follifields ( 1 0 0 ~ with cle,I3 and 50 C-cells in at least one low power field ( ~ O O X ) .This ~ ~ problem has been further complicated by the lumping of biologically different C-cell proliferations under the umbrella designation of CCH. Our results support the hypothesis that physiologic CCH should be separated from neoplastic CCH because of morphologic differences. A comparison of these two forms of CCH, based o n our findings and those of others, is compiled in Table 2. Physiologic CCH was defined using previously published criteriaz5which consisted of identifying at least 50 calcitonin positive cells in one low power field ( 1 0 0 ~ ) . Although arbitrary, this number of cells is greater than that found in most normal subjects and has given rise to hypercalcitoninemia. In this study, these hyperplastic Ccells could not be recognized with certainty on H & E sections because of their morphologic similarities to follicular cells and histiocytes. Our cases were associated with follicular thyroid neoplasms and Hashimoto's thyroiditis. In fact, we have now seen physiologic CCH in 2 out of 28 cases (7%) of Hashimoto's thyroiditis stained with calcitonin by IHC. One of these was previously described.I9 Other reported associations include hyperparathyroidism,",'" previous hemithyroidectomy,*'' advanced age,'53'band non-Hodgkin's lymphoma of the thyroid.21 Mechanisms involved in the development of this form of 754 CANCER February 15, 1996 / Volume 77 I Number 4 TABLE 2 Physiologic vs. Neoplastic CCH Biologic variables Association with MEN IUFMTC Hypercalcitoninemia Horntonally induced Genetic defects of chromosome 10 Physiologic CCH Neoplastic CCH No Yes Probably Probably not Yes Yes NO Yes Histologic variables ~ Detectable on H&E Cytologic atypia Seen adjacent to MTC Bilaterality Staining with NCAM-MoAB735 (41) Calcitonin reactivity No No No Usually nota No Yes ~~ Yes Yes Yes Yes Yes Yes "Only one case was bilateral. NCAhl-MoAb: neural cell adhesion molecule nioiioclonal antibody. CCH are not well established. TSH overstimulation is a suspected cause in neonates and in some cases of hypothyroidism. Physiologic CCH has been induced in laboratory animals via TSH overstiniulation.'"-"2 However, other factors must also be important since, in our experience, most patients with physiologic CCH do not have elevated serum TSH levels. It is obvious that much remains to be learned about the interactions between C-cells and follicular cells. These interactions might explain the great variations in the number of C-cells under physiologic and pathologic conditions. In experimental animals, it has been suggested that C-cells may play a role in the modulation of thyroid function by paracrine mechanisms.3j Neoplastic CCH is the precursor of MTC and is most commonly seen in patients with the familial forms of the disease including MEN IIa, MEN IIb, and FMTC.""13 All three of these familial disorders are associated with genetic defects in the paracentromeric region of chromosome 10.34-4" Therefore, it is suspected that they may simply represent variants of the same disease. The gene responsible for MEN Ha, IIb, and FMTC has recently been identified as the tyrosine receptor proto-oncogene known as R E T . ~ " - ~ ~ Unlike physiologic CCH, neoplastic CCH is associated with cytologic atypia and is, therefore, recognizable on routine H & E sections. Because of this feature, numerical criteria are irrelevant and unnecessary for the diagnosis. After all, neoplastic transformation is a qualitative, not a quantitative, change. In fact, these atypical cells resemble those of MTC both morphologically and antigenically. In addition, a recent study has shown that both MTC and neoplastic CCH stain positively with the recently described neural cell adhesion molecule (NCAM) monoclonal antibody 735, whereas physiologic CCH gives negative results (Table 2).43NCAM is a member of the immunoglobulin family of membrane receptors implicated in cell adhesion. This molecule is found in a variety of endocrine tumors and, therefore, is not a specific marker of C-cells. Carney et al.' were correct in pointing out the inaccuracy of the term CCH in describing the precursor lesion of MTC in MEN 11. C-cell dysplasia or medullary carcinoma in situ would be more biologically correct. In this study, we refer to it as "neoplastic" CCH despite its contradictions because we realize the difficulties in changing well established terminology. During our review of medullary carcinomas, an example of neoplastic CCH was found adjacent to a small sporadic MTC found incidentally in a laryngectomy specimen. This case, which was not included in the present series, was unusual in that the tumor was only 0.4 cm in greatest dimension. It should be noted that some patients with medullary carcinomas thought to be sporadic, show germ line RET proto-oncogene mutations, indicating that their tumors are in fact familial.42Likewise, 11 of the 12 cases of familial MTC described in this study had adjacent neoplastic CCH and most of the tumors were small (0.1 cm to 1.7 cm, mean 0.7 cm, median 0.4 cm). The remaining case of familial MTC was associated with a 3.0 cm tumor and no adjacent CCH was found. Therefore, we believe that neoplastic CCH is the precursor of both sporadic and familial MTC and that tumors of relatively large size before detection (most sporadic cases) have most likely overgrown the neoplastic CCH. If this were the case, it would not be surprising that most cases of neoplastic CCH would be seen in familial MTC where multifocality is common and screening tests often result in detection of small or microscopic tumors. A recent study of proliferative activity in familial cases of CCH and MTC, measured by proliferating cell nuclear antigen (PCNA) and nucleolar organizing regions silver staining (AgNOR), has shown a higher proliferative activity in medullary carcinoma as compared to CCH.44This data also supports the contention that CCH is the precursor lesion of MTC in familial cases. Whereas most publications on CCH have focused on the neoplastic form, we believe that physiologic CCH is probably more common. Also, it is not known how often physiologic CCH is associated with hypercalcitoninemia because serum calcitonin is usually not measured in these cases. In this study, 3 cases demonstrated hypercalcitoninemia, and the remaining 16 either had no serum calcitonin measurements or the information was unavailable. Barbot et a1.2Ddemonstrated hypercalcitoninemia and CCH in 3 out of 24 patients with Hashimoto's thyroiditis. The important implication is that a patient with a thyroid mass and hypercalcitoninemia may have something other than MTC (i.e., follicular neoplasm, Hashi- C-Cell HyperplasialPerry et al. moto’s thyroiditis, etc.). The natural history of physiologic CCH is not known. Due to its reactive nature, it seems logical to assume that it is a reversible condition. This is in sharp contrast to the malignant potential of familial CCH which usually progresses to medullary carcinoma, further emphasizing the importance of separating physiologic from neoplastic CCH. In conclusion, we have demonstrated that physiologic and neoplastic CCH are not only biologically distinct entities, they are also morphologically distinct. Therefore, different diagnostic criteria should be utilized for each one. We believe that neoplastic CCH is the precursor of familial and, most likely, sporadic MTC whereas physiologic CCH is a reactive process not associated with MTC. Pathologists and clinicians should be aware of these distinctions and not lump the two entities under the same terminology. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Pearse AGE. The cytochemistry of the thyroid C-cells and their relationship to calcitonin. Proc R SOCLond B Biol Sci 1966; 164:478-87. Hazard JB, IHawk WA, Crile G. Medullary (solid) carcinoma ol‘ the thyroid: a clinicopathologic entity. J Cliri Er~docririol Metab 1959; 191152-61. Nbores-Saavedra J, LiVolsi VA, Williams EL). Medullary carcinoma. Sernin Diagr7 Patliol 1985;2:137-46. Al Saadi AA. Ultrastructure in C-cell hyperplasia in asymptcimatic patients with hypercalcitoninemia and a family history of medullary thyroid carcinoma. Hum Pafhol 1981; 1it1617 -22. Biddinger PW, Ray M. Distribution of C-cells in the normal and diseased thyroid gland. Pathol A m i i 1993;28:205-29. Carney JA, Sizemore GW, Hales AB. Multiple endocrine neoplasia, type 2b. Pafhobiol Aiinzt 1978;8:105-53. DeLellis RA, Wolfe HI. The pathobiology of the human calcitonin (C)-ceIl; a review. Pafhol Aizriii 1981; 16:25-52. Ekblom M, Valimaki M, Pelkonen R, Jansson R, Sivula A, Franssila K. Familial and sporadic medullary thyroid carcinoma: clinical and irnmunohistocheniical findings. QJ Med 1!387;65:899-910. Hazard JB. The C-cells (parafollicular cells) of the thyroid gland and medullary thyroid carcinoma. An7 1 Pathol 1!377;88:2 14-49. Mendelsohn G, Eggleston JC, Weisburger WR, Gann DS, Baylin SB. Calcitonin and histaminase in C-cell hyperplasia and medullary thyroid carcinoma. A light microscopic and imniunohistoc:hemical study. A m J Patliol 1978;9235-52. Sinipson WJ,Carruthers JS, Malkins D. Results of a screening program for C-cell disease (medullary thyroid cancer and C-cel! hvDerD1asia). Cancer 1990;G5:1570-6. . Ulribe M, Grimes M. Fenoglio-Preiser CM, Feind C. Medullary carcinoma of the thyroid gland. A m J Sztrg Pnflzol 1985;9:577--94. Williams El), Ponder BJ, Craig RK. Immunohistochemical study of calcitonin gene-related peptide in human medullary carcinoma and C cell hyperplasia. Clin Etzdocrinol 1987;27:107- 14. Wolfe HJ, Melvin KEW, Cervi-Skinner SJ. Al Saadi AA, Julian JF. Jackson CE, Tashjian AH Jr. C-cell hyperplasia preceding ,. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 755 medullary thyroid carcinoma. N Eiigl J Med 1973;289:43741. Gibson WGH. Peng TC, Croker BP. C-cell nodules in adult human thyroid. A common autopsy finding. Arii J Cliii Patliol 1981;75:347-50. Gibson WGH, Peng TC, Croker BP. Age-associated C-cell hyperplasia in the human thyroid. An7 J Patliol 1982; 106~388-93. Tomita T, Millard DM. C-cell hyperplasia in secondary hyperparathyroidism. Histopathology 1992;2 1:4G9- 74. Ljungberg 0. Dymling JF. Pathogenesis of C-cell neoplasia in thyroid gland. Acta Pnthol Microbial Scnnd A 197280577-88. Albores-Saavedra 1. C-cell hyperplasia [letter].Am J Sirrg Patho1 1989; 13:987. Barbot N, Guyetant S, Beldent V, Akrass A, Cerf 1, Perdrisot R, et al. Chronic autoimmune thyroiditis and C-cell hyperplasia. Study of calcitonin secretion in 24 patients. Aiiri Elidocrinol (Paris). 1991; 52(2):109- 12. Baschieri L, Castagna M, Fierabracci A, Antonelli A. Del Guerra P, Squartini F. Distribution of calcitonin- and somatostatin-containing cells in thyroid lymphoma and in Hashirnoto’s thyroiditis. Appl P a t h 1 1989;7:99- 104. Biddinger PW, Brennan MF, Rosen PP. Symptomatic C-cell hyperplasia associated with chronic lyinphocytic thyroiditis. A m J Sitrg Pafliol 1991; 15599-604. Libbey NP, Nowakowski KJ, Tucci JR. C-cell hyperplasia of the thyroid in a patient with goitrous hypothyroidism and Hashimoto’s thyroiditis. J Sitrg Pntliol 1989; 13:71-7. Ulbright TM, Kraus FT, O’Neal LW. C-cell hyperplasia developing in residual thyroid following resection for sporadic medullary carcinoma. Cancer 1981;48:2076-9. Albores-Saavedra 1, Monforte H, Nadji M, Morales AR. Ccell hyperplasia in thyroid tissue adjacent to follicular cell tumors. Hitiri Pafliol 1988;19:795-9. Scopsi L, DiPalma S, Ferrari C, Holst J J , Rehfeld JF, Rilke F. Ccell hyperplasia accompanying thyroid diseases other than medullary carcinoma: an irnmunoc~ochemicalstudy by means of antibodies to calcitonin and sornatostanin. hlod Path01 1991;4:297-304. LiVolst VA, Feind CR, LoGerfo P, Tashjian AH Jr. Demonstration by immunoperoxidase staining of hyperplasia of parafollicular cells in the thyroid gland in hyperparathyroidism. J Clin Endocrim1 Metab 1973;37:550-59. Aliapodios MA, Kacoyaiiis GP. Thyroid stimulating hormone and thyrocalcitonin content in rat thyroid. J Sitrg Res 1976;2 1:449-52. Clark OH, Rehfeld SJ, Castner B, Stroop J, Loken HF, Deftos LJ. Iodine deficiency produces hypercalcemia and hypercalcitoninemia in rats. Surgery 1978;831626-32. Peng TC, Cooper CW, Garner SC, Volpert EM. Hypercalcitonisni and C-cell hyperplasia in rats with goiters produced by a low iodine diet or propylthiouracil. Plzanizncol Exp 1 ’ 1 ~ 1978;206:710-7. Peng TC, Cooper CW, Petrusz P, Volpert EM. Identification of C-cells in normal and goitrous rat thyroid tissues using antiserum to rat thyrocalcitonin and the immunoperoxidase bridge technique. Etidocrinobgy 1975;97:1537-41. Yasumura S, Burk M, Chausmer A, Mittleman R. Wallach S. Thyroidal content of thyrocalcitonin in hypothyroid and hyperthyroid rats. Eizdocrinology 1967;81:256-60. Gkonos PJ, Tavianini MA, Liu CC, Roos BA. Thyrotropinreleasing hormone gene expression in normal thyroid parafollicular cells Mol Endocriiiol 1989;3:2101-9. 756 CANCER February 15,1996 / Volume 77 / Number 4 34. Shimotake T, Iwai N , Yanagihara J, Tikiwa K, Tamapa N, Yaniamoto M, er al. Prediction of affected MEN 2A gene carriers by DNA linkage analysis for early total thyroidectorny: a progress in clinical screening program for children with hereditary cancer syndrome. J Pediatr Surg 1992; 27:444-6. 35. Howe JR, Lairmore TC, Mishra SK, Dou S, Veile R, Wells SA Jr., et al. Improved predictive test for MEN 2, using flanking dinucleotide repeats and RFLPs. Am J H u m Genet 1992;51:1430-42. 36. Lairmore TC, Howe JR, Korte JA, Dilley WG, Aine L, Aine E, et al. Familial medullary thyroid carcinoma and multiple endocrine neoplasia type 2b map to the same region of chromosome 10 as multiple endocrine neoplasia type 2a. Geno/?Z~CS 1991;9281-92. 37. Mulligan LM, Gardner E, Smith BA, Mathew CGP, Ponder BAJ. Genetic events in tumour initiation and progression in multiple endocrine neoplasia type 2. Genes Cliromosom Cancer 1993;6:166-77. 38. Narod SA, Sobol H, Schuffenecker I, Lavorre MF, Lenoir GM. The gene for MEN 2A is tightly linked to the centromere of chromosome 10. Hum Genet 1991;88:529-30. 39. Thakker RV. The molecular genetics of the multiple endocrine neoplasia syndromes. C h i Etadocrinol 1993;38: 1- 14. 40. Mulligan LM, Kwok JBJ, Healey CS, Elsdon MI, Eng C, Gardner E, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993; 363:458-60. 41. Lips CJM, Landsvater RM, Hoppener JWM, Geerdink RA, Blijham G, Jansen-Schillhorn von Veen JM, et al. Clinical screening as compared with DNA analysis in families with multiple endocrine neoplasia type 2A. N En@ J Med 1994;331:828-35. 42. Komminoth P, Kunz EK, Matias-Guiu X, Hiort 0, Christiansen G, Colomer A, et al. Analysis of RET protooncogenepoint mutations distinguishes heritable from nonheritablemedullary thyroid carcinomas. Cancer 1995;76:47989. 43. Komminoth P, Roth 1, Saremaslani P, Matias-Guiu X, Wolfe HJ, Heitz PU. Polysialic acid of the neural cell adhesion molecule in the human thyroid: a marker for medullary thyroid carcinoma and primary C-cell hyperplasia. Am J Surg Pudiol 1994; 18~399-411. 44. Matias-Guiu X, Peiro G, Esquius J , Oliva E, Cabezas R, Colomer A, et al. Proliferative activity in C-cell hyperplasia and medullary thyroid carcinoma. Evaluation by PCNA immunohistochemistry and AgNORs staining. Purhol Res Prnct 1995; 191~42-7.