estrogen receptors and cathepsin d in human thyroid tissue

1991

Estrogen Receptors and Cathepsin D in Human Thyroid Tissue Thiery Mifaye, Ph.D.,* Christine Millet, M.Pharm.,* Jean-Louis Kraimps, M.D.,t Brigitte Aubouin, M.S.,* Jacques Barbier, M.D.,t and Franqois Begon, M.D.*

Background. To investigate the significance of estrogen receptors (ER) in the pathogenesis of thyroid dyspla- sia, the authors analyzed, by analogy with breast cancers, ER and three estrogen-regulated proteins: progesterone receptor (PR), cathepsin D, and pS2 protein, in cytosols of 42 human thyroid tissues.

Methods. ER and PR were measured by an immuno- enzymatic assay and cathepsin D and pS2 by an immunoradiometric assay. Tissue specimens included 7 normal tissues, 6 benign nodules, 8 toxic adenomas, 7 from patients with Graves disease, and 14 carcinomas.

Results. ER was present at very low concentrations, with no statistical difference between neoplastic and nonneoplastic tissues. The mean levels of cathepsin D, expressed as pmol/mg protein minus thyroglobulin, were higher in the 14 carcinomas ( P = 0.0003), the 7 specimens from patients with Graves disease (P = 0.006), and the 8 toxic adenomas (P = 0.04) than in the 7 normal thyroid tissues. A significant difference also was observed between the carcinomas ( P = 0.003) and six benign nodules. Compared to TNM parameters, cathepsin D concentrations correlated with tumor size: higher cathepsin D levels were found in pT4 than in pT2 and pT3 carcinomas. All the tissues tested were negative for PR and pS2 protein.

Conclusions. The results clearly indicate a significant difference between neoplastic and normal thyroid tissue in terms of the amount of cathepsin D, but not that of ER. This suggests that cathepsin D probably is not regulated by estrogen but simply is a marker of protease activity during invasion by thyroid carcinomas. Cancer 1993; 72~1991-6.

Key words: estrogen receptors, cathepsin D, progesterone receptors, pS2, thyroid neoplasms, goiter, adenoma.

From the Departments of *Biophysics and tSurgery, Jean Ber- nard Hospital, Poitiers, France.

Address for reprints: Thieny Metaye, Ph.D., Groupe de Re- cherche en Endocrinologie Experimentale et Clinique, Laboratoire de Biophysique, Hbpital Jean Bernard, BP 577, 86021 Poitiers Cedex, France.

Accepted for publication April 26, 1993.

The presence of a specific hormone receptor in a cancerous tissue indicates that the hormone may influence cell multiplication and that the tumor may be hormone de- pendent.' Immunohistochemica12-4 and biochemical studies5-13 have established that thyroid tissues, including carcinomas, possess estrogen (ER), progesterone (PR), and androgen receptors. This, together with the higher incidence of thyroid cancers in women than in men, suggested a role of sex hormones in tumor initiation. Quantitative measurement^,^-^,'^ however, showed low steroid receptor expression in the thyroid gland, raising questions as to their function. The synthesis of several proteins is regulated by estrogen in breast cancers, and such proteins thus could serve as comple- mentary markers of an estrogen responsiveness. PR is the best example of estrogen-regulated protein in breast cancer,I4 and is associated with an improved prognosis. Cathepsin D is a lysosomal acid aspartyl protease regulated by progesterone in human end~metrium'~ and by estrogen in human breast cancer cells.16 pS2 protein was shown to be a secretory protein in breast cancer cells, being predominantly expressed in ER+ breast tumors and virtually absent from ER- breast tumors." Furthermore, similar levels of pS2 mRNA expression have been observed in breast carcinomas and in malignant and benign thyroid tumors.'*

We measured ER, PR, cathepsin D, and pS2 protein in cytosols of various thyroid tissues to determine the potential role of ER in malignant transformation.

Materials and Methods

Thyroid Samples

ER, PR, cathepsin D, and pS2 protein were analyzed in cytosols of 42 human thyroid specimens, including 7 normal tissues, 6 benign nodules, 8 toxic adenomas, 7 from patients with Graves disease, and 14 carcinomas. All specimens were obtained at surgery and stored in liquid nitrogen.

1992 CANCER September 15, 1993, Volume 72, No. 6

Table 1. Characteristics of 14 Patients With Thyroid Carcinomas

pTNM classification

Age (yr) Sex Histologic type PT PN PM Ba 82 F Papillary 4 l a 0 RO 39 F Papillary 2 0 0 Pi 36 F Papillary 2 0 0 Ga 28 F Papillary 4 l a 0 Bu 59 F Papillary 4 l b 0 Bo 25 M Papillary 2 la 0 Pr 66 M Pap i 11 a r y 4 1 0 Ca 80 F Anaplastic 4 0 0 P1 67 M Anaplastic 4 l b 0 Rh 79 F Anaplastic 4 l b 0 De 68 F Follicular 3 0 1 Ch 47 F Follicular 2 l b 1

s c 55 F Tra becular 4 l a 1 Pa 51 M Medullary 4 l b 1

The age and sex of the 14 patients with carcinomas, together with their histologic diagnosis and pTNM classification (UICC recommendations), are shown in Table 1.

Processing of Thyroid Tissues

Frozen thyroid tissues were mechanically pulverized in an impact grinder at -180°C (Freezer mill; SPEX In- dustries, Inc., Edison, NJ), then homogenized at 4°C in three volumes of Tris buffer (lo-* mol/l Tris-HC1, 1.5 X mol/l ethylene diamine tetraacetic acid, 1.2 X

mol/l dithiotreitol, and lo-’ mol/l sodium molyb- date, pH 7.4) using an Ultra-Turrax homogenizer (Osi, Paris). The homogenate was centrifuged at 105,000 X g at 4°C for 1 hour (Beckman L8-55; Beckman Jnstru- ment, Inc., Fullerton, CA) to isolate the cytosol. The same freshly prepared cytosol samples were used to determine ER, PR, cathepsin D, and pS2 protein.

ER and PR Assays

The enzyme immunoassay (EIA) kits for ER and PR were purchased from Abbott Laboratories (Rungis, France) and used according to the manufacturer’s instructions. EIA is based on a “sandwich” technique in which receptors bind a monoclonal antibody-coated bead; measurement is made using a second anti-receptor monoclonal antibody conjugated to horseradish peroxidase. Results are expressed as femtomoles of receptor per milligram of total protein minus thyroglobulin (fmol/mg Pt - Tg). This expression was preferred to fmol/mg total protein because the major portion of thyroglobulin is not intracellular and the different thyroid tissues did not contain the same amount of thyroglobu-

lin. The detection limit of the receptor assay was established from the 95% confidence interval of standard zero. Protein concentrations were determined using the method of Lowry et aI.l9

Cathepsin D Assay

The total amount of cathepsin D (52-, 48-, and 34-kilo- dalton species) was determined in a solid-phase, two- site immunoradiometric assay using an ELSA-cath D kit (CIS Bio-Industries, Gif-sur-Yvette, France). The assay involves two monoclonal antibodies, one coated on the solid phase and the other labeled with iodine 125. Cy- tosols were diluted to 1 mg/ml of protein with Tris buffer containing 10% glycerol, and the assay was performed according to the manufacturer’s instructions. Results are expressed as pmol/mg Pt - Tg. The detection limit of the cathepsin D assay was established from the 95% confidence interval of standard zero.

pS2 Protein Assay

We used a one-step, double-determinant, solid-phase immunoradiometric assay (ELSA-pS2 kit; CIS Bio-In- dustries) according to the manufacturer’s instructions. Results are expressed as ng/mg Pt - Tg. The detection limit of the pS2 assay was established from the 95% confidence interval of standard zero.

Thyroglobulin Assay

Cytosolic thyroglobulin was determined using an immunoradiometric assay kit (Henning Berlin GmbH, Berlin, Germany), which is based on two monoclonal antibodies, one coated directly on the tube and the

ER and Cathepsin D in Thyroid Tissue/M&aye' et al . 1993

Table 2. Estrogen Receptor and Cathepsin D Values in Thyroid Carcinoma Cytosols

Estrogen receptors Cathepsin D Patients (fmol/mgPt-Tg) (pmol/mgPt-Tg)

Primary tumors Ba 3.9 39

Ro 3.0 20 Pi 2.0 15 Ga 1.9 31 Ca ND 28 P1 1.8 26 Rh 2.8 49 De ND 16 Ch 1.3 14 Pa ND 17 s c 1.7 64

metastases Bu 1.4 44 Bo 8.4 34 Pr 0.7 33

Lymph node

ND: not detected.

other labeled with iodine 125. The dilutions of cytosols were performed with standard zero.

Statistical Methods

Differences between cathepsin D values in the various thyroid tissues were analyzed using the Kruskall- Wallis and Mann-Whitney nonparametric tests. An associa- tion between cathepsin D and tumor size was sought using the Spearman rank correlation coefficient.

Results

ER and cathepsin D were determined in human thyroid carcinomas (1 1 primary tumors and 3 lymph node metastases; Table 2). Eight of the primary tumors were positive for ER (1.3-3.9 fmol/mg Pt - Tg), whereas three were below the detection limit. In agreement with Chaudhuri et al.," all the papillary carcinomas were positive for ER, whereas the medullary carcinoma was negative. Cathepsin D values ranged between 14 and 64 pmol/mg Pt - Tg in primary tumors with high values (>30 pmol/mg Pt - Tg) observed in all histologic types, with the exception of follicular and medullary carcinomas, few of which were tested. In lymph node metastases, ER values ranged from 0.7-8.4, and cathepsin D values from 33-44 pmol/mg Pt - Tg.

Mean ER and cathepsin D values in thyroid carcinomas were compared to those in normal tissues, benign nodules, toxic adenomas, and Graves disease (Ta- ble 3). The cathepsin D content of the carcinomas was

significantly different from that of the normal tissues (P = 0.0003) and benign nodules (P = 0.003), but not from that of the other two nonneoplastic thyroid tissues. Cathepsin D values in the toxic adenomas and Graves disease samples ranged from 7-81 and 10-31 pmol/mg Pt - Tg, respectively, with a mean value statistically different from that of normal tissues. Moreover, cathepsin D values were 40 and 68 pmol/mg Pt - Tg in two toxic adenomas, compared to 8 and 12 pmol/mg Pt - Tg in adjacent normal tissues, showing the specificity of cathepsin D expression in the pathologic tissues.

The number of specimens positive for ER was five of seven normal tissues, two of six benign nodules, six of eight toxic adenomas, and six of seven Graves disease. There was no significant difference in ER content between neoplastic and nonneoplastic tissues. PR and pS2 protein were not detectable in any of the tissues tested.

Cathepsin D levels in primary tumors were significantly associated with tumor size (Fig. 1). The seven carcinomas classified as pT4 showed cathepsin D values from 17-64 (mean, 36), the three classified as pT2 had values of 14-20 (mean, 16) and the one classified as pT3 had a value of 16 pmol/mg Pt - Tg. No correlation was found with other parameters such as age, sex, or metastatic potential.

Discussion

Steroid receptors are present in normal and pathologic thyroid ER and PR were identified later in the nucleus of follicular cells by means of immunohistochemical techniques,'-* raising doubts as to the exact receptor-containing cell types. Our results confirm the presence of ER in cytosols of normal human thyroid

TabIe 3. Mean Estrogen Receptor and Cathepsin D Values in Normal and Pathologic Thyroid Tissue

Estrogen receptors Cathepsin D (mean t SD) (mean f SD)

Normal tissue (7 cases) 2.1 f 1.4 9.8 f 2.4

Toxic adenoma (8 cases) 1.1 f 0.8 33.2 k 27.8* Graves disease (7 cases) 1.1 k 0.5 20.3 k 7 3 Carcinoma (14 cases) 2.1 k 2.1 30.7 & 14.6$.§

SD: standard deviation. * Significantly different from normal tissue (P = 0.04). t Significantly different from normal tissue (P = 0.006). $ Significantly different from normal tissue ( P = 0.0003). 5 Significantly different from benign nodule (P = 0.003). Estrogen receptor and cathepsin D values are expressed as fmol/mgI't-Tg and pmol/mgPt-Tg, respectively. Statistical differences between the groups were analysed by nonparametric Kruskall-Wallis and Mann-Whitnev tests.

Benign nodule (6 cases) 1.8 f 3.2 11.2 f 7.3

1994 CANCER September 15,1993, Volume 72, No. 6

70 -

60.

50.

40.

30.

20.

10,

, Ca hepsin D (p..l/rng Pt -Tg)

I 4

Pm Pm PT4 Tumor size

Figure 1. Correlation between cathepsin D values in primary tumors and tumor size according to the pTNM classification of thyroid carcinomas. The Spearman rank correlation coefficient is rs = 0.77, with P = 0.006 (statistically significant).

glands and a variety of pathologic thyroid tissues, including benign nodules, toxic adenomas, Graves disease, and carcinomas. In contrast to some report^,'^,",'^ PR was not detectable in any of the tissues tested.

The EIA for ER was preferred to the classic hormone-binding assay principally because of the excel- lent correlation reported between the two,” and the better reproducibility, sensitivity, and accuracy of the former for the determination of low ER concentrations.’l As previously rep~r ted ,~-~ , ’~ we found a low ER content in all the thyroid tissues; however, no significant difference was observed between neoplastic and nonneoplastic tissues. This apparent discrepancy with the reports by Clark et al.,’ Chaudhuri et al.,” and Miki et may stem from the assay methods (hormone- binding assay versus EIA) and the expression of the results (fmol/mg of total protein versus fmol/mg of total protein minus cytosolic thyroglobulin). Indeed, we found higher thyroglobulin concentrations in cytosols of the nonneoplastic tissues than those of the carci-

nomas (data not shown), indicating that expression per milligram of total protein would lead to an underesti- mation in normal thyroid tissue.

Although thyrotropin, epidermal growth factor, and insulin-like growth factor are the main compounds influencing the growth of normal and neoplastic thyroid tissue,” several lines of clinical evidence suggest that estrogen may affect the initiation’,23 and prognosis of thyroid cancer^.'^-^^ The low ER content in thyroid tissues, however, raises questions as to the function of this receptor, which may be answered by measuring estrogen-regulated proteins. We found significantly elevated levels of cathepsin D in pathologic thyroid tissues from patients with toxic adenoma, Graves disease, and carcinoma. By analogy with breast cancers, these results may indicate a role of ER in the regulation of cathepsin D levels in thyroid carcinomas. It is possible, however, that synthesis of estrogen-regulated proteins is organ specific, and that cathepsin D expression is re- lated only to neoplastic transformation of the thyroid. Furthermore, we failed to detect pS2, another estrogen- regulated protein in breast cancer.

Many normal and cancerous human cells (mela- noma and lung, stomach, colon, kidney, breast, ovary, bladder, pancreas, liver, and neural carcinomas) contain cathepsin D.27-29 Elevated cathepsin D content in human thyroid lesions has not been reported, although increased acid protease activities have been found in Graves disease,30 toxic a d e n ~ m a s , ~ ~ and papillary carci- n o m a ~ . ~ ~ The role of cathepsin D in the thyroid is not clear, but this lysosomal acid endopeptidase has been described as the main enzyme involved in thyroglobulin p r o t e ~ l y s i s . ~ ~ , ~ ~ Furthermore, cathepsin D appears to be quantitatively more important than thiol proteases in the initial phase of the dige~tion?~ and its activity is stimulated by thy r~ t rop in .~~”~ This may explain why hyperfunctional thyroids in patients with toxic adenoma and Graves disease, for example, contain elevated cathepsin D levels.

The possible relationship between neoplastic transformation and increased proteolytic activity of malignant cells has been studied e~tensively.~’ Proteases, including cathepsin D, play a role in metastatic dissemina- tion and contribute to basement membrane and connective tissue degradation, allowing vascular endo- thelial crossover. High concentrations of cathepsin D have been found in poor-prognosis primary breast tu- m o r ~ , ~ ~ and are associated with high myomehial inva- sive potential in human endometrial cancers.15

In thyroid carcinoma^,^^^^^^^^ tumor invasion beyond the thyroid capsule is an important predictor of recurrence and death. Therefore, the higher levels of cathepsin D in pT4 carcinomas (Fig. 1) than in Stages

ER and Cathepsin D in Thyroid Tissue/Mitayi et al. 1995

pT2 and pT3 make cathepsin D a potential marker of poor prognosis in thyroid carcinomas.

In conclusion, the simultaneous presence of ER and cathepsin D in thyroid cancers could indicate, by analogy with breast cancers, a physiologic role of ER in the regulation of cathepsin D expression. A significant difference between neoplastic and normal thyroid tissue, however, was observed in terms of cathepsin D content but not ER content. This indicates that cathepsin D probably is not regulated by estrogen but simply is a marker of protease activity during tumor invasion. Al- though increased cathepsin D levels have been correlated, in this study, with thyroid carcinoma size, further studies of larger populations will be necessary to determine the potential prognostic value of cathepsin D in thyroid carcinomas.

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

Henderson BE, Ross RK, Pike MC, Casagrande JT. Endogenous hormones as a major factor in human cancer. Cancer Res 1982;

Money SR, Muss W, Thelmo WL, Boeckl 0, Pimp1 W, Kaindl H, et al. Immunocytochemical localization of estrogen and progesterone receptors in human thyroid. Surgery 1989; 106:975-9. Hiasa Y, Nishioka H, Kitahori Y, Yane K, Nakaoka S, Ohshima M, et al. lmmunohistochemical detection of estrogen receptors in paraffin sections of human thyroid tissues. Oncology 1991; 48:421-4. Diaz NM, Mazoujian G, Wick MR. Estrogen-receptor protein in thyroid neoplasms: an immunohistochemical analysis of papillary carcinoma, follicular carcinoma, and follicular adenoma. Arch Pathol Lab Med 1991; 115:1203-7. Keshgegian AA, Wheeler JE. Estrogen receptor protein in malignant carcinoid tumor: a report of 2 cases. Cancer 1980; 45:293-6. Stedman KE, Moore GE, Morgan RT. Estrogen receptor proteins in diverse human tumors. Arch Surg 1980; 115:244-8. Molteni A, Warpeha RL, Brizio-Molteni L, Fors EM. Estradiol receptor-binding protein in head and neck neoplastic and normal tissue. Arch Surg 1981; 116:207-10. Clark OH, Gerend PL, Davis M, Goretzki PE, Hoffman PG Jr. Estrogen and thyroid-stimulating hormone (TSH) receptors in neoplastic and nonneoplastic human thyroid tissue. 1 Surg Res

Hampl R, Nemec J, Heresova J, Kimlova I, Starka L. Estrogen receptors in human goitrous and neoplastic thyroid. Endocrinol

Schaefer CJ, Geelhoed GW, Dadu P. Thyroid disorders and steroid receptor proteins. A m Surg 1986; 52:514-8. Chaudhuri PK, Patel N, Sandberg L, Prinz RA. Distribution and characterization of steroid hormone receptors in human thyroid tissue. World 1 Surg 1986; 10:737-44. Chaudhuri PK, Prinz R. Estrogen receptor in normal and neoplastic human thyroid tissue. A m 1 Otolaryngol 1989; 10:322-6. Miki H, Oshimo K, Inoue H, Morimoto T, Monden Y . Sex hormone receptors in human thyroid tissues. Cancer 1990;

42~3232-9.

1985; 38239-96.

EXP 1985; 191227-30.

66: 1759-62.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

Horwitz KB, McGuire WL. Estrogen control of progesterone receptor in human breast cancer: correlation with nuclear processing of estrogen receptor. ] Biol Chem 1978; 253:2223-8. Maudelonde T, Martinez P, Brouillet J-P, Laffargue F, Pages A, Rochefort H. Cathepsin-D in human endometrium: induction by progesterone and potential value as a tumor marker. J Clin Endocrinol Metab 1990; 70:115-21. Veith FO, Capony F, Garcia M, Chantelard J, Pujol H, Veith F, et al. Release of estrogen-induced glycoprotein with a molecular weight of 52,000 by breast cancer cells in primary culture. Cancer Res 1983; 43:1861-8. Rio MC, Bellocq JP, Gairard B, Rasmussen UB, Krust A, Koehl C, et al. Specific expression of the pS2 gene in subclasses of breast cancers in comparison with expression of the estrogen and progesterone receptors and the oncogene ERBB2. Proc Nut1 Acad Sci

Zaretsky JZ, Weiss M, Tsarfaty 1, Hareuveni M, Wreschner DH, Keydar I. Expression of genes coding for pS2, c-erbB2, estrogen receptor and the H23 breast tumor-associated antigen: a compar- ative analysis in breast cancer. FEBS Lett 1990; 265:46-50. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with Folin phenol reagent. 1 Biol Chem 1951; 193:265- 75. Leclercq G, Bojar H, Goussard J, Nicholson RI, Pichon M-F, Piffanelli A, et al. Abbott monoclonal enzyme immunoassay measurement of estrogen receptors in human breast cancer: a European multicenter study. Cancer Res 1986; 46(Suppl):4233s- 6s. Thorpe SM. Monoclonal antibody technique for detection of estrogen receptors in human breast cancer: greater sensitivity and more accurate classification of receptor status than the dextran- coated charcoal method. Cancer Res 1987; 476572-5. Bidey SP. Control of thyroid cell and follicle growth: recent ad- vances and current controversies. Trends Endocrinol Metab 1990;

Paloyan E, Hofmann C, Prinz RA, Oslapas R, Shah KH, KU WW, et al. Castration induces a marked reduction in the incidence of thyroid cancers. Surgery 1982; 92:839-48. Tubiana M, Schlumberger M, Rougier P, Laplanche A, Benha- mou E, Gardet P, et al. Long-term results and prognostic factors in patients with differentiated thyroid carcinoma. Cancer 1985;

Simpson WP, McKinney SE, Carruther JS, Gospodarowicz MK, Sutcliffe SB, Panzarella T. Papillary and follicular thyroid cancer: prognostic factors in 1,578 patients. A m 1 Med 1987;

Cady B, Sedgwick CE, Meissner WA, Wool MS, Salzman FA, Werber J. Risk factor analysis in differentiated thyroid cancer. Cancer 1979; 43:810-20. Reid WA, Valler MJ, Kay J. Immunolocalisation of cathepsin D in normal and neoplastic human tissues. 1 Clin Pathol 1986;

Parham D, Whitaker JN, Berard CW. Cellular distribution of cathepsin D in childhood tumors. Arch Pathol Lab Med 1985;

Maguchi S, Taniguchi N, Makita A. Elevated activity and increased mannose-6-phosphate in the carbohydrate moiety of cathepsin D from human hepatoma. Cancer Res 1988; 48:362-7. Yoshinari M, Inoue K, Nakashima T, Okamura K, Shiroozu A, Nishitani H, et al. Acid protease activity in thyroid gland from patients with Graves' disease. Metabolism 1983; 32:348-54. Yoshinari M, Okamura K, Shiroozu A, Nakashima T, Inoue K,

USA 1987; 84~9243-7.

1 : 174-8.

55:794-804.

83:479-88.

39~1323-30.

109:250-5.

1996 CANCER September 15, 2993, Volume 72, No. 6

Fujishima M. Acid thyroglobulin protease activities in human diseased thyroid glands. Clin Endocrinol 1987; 26:685-91.

32. Sinadinovic J, Cvejic D, Savin S, Micic JV, Jancic-Zguricas M. Enhanced acid protease activity of lysosomes from papillary thyroid carcinoma. Cancer 1989; 63:1179-82.

33. Balasubramaniam K, Deiss WP Jr. Characteristics of thyroid lyso- soma1 cathepsin. Biochim Biophys Acta 1965; 110:564-70.

34. Dunn AD, Dunn JT. Thyroglobulin degradation by thyroidal proteases: action of purified cathepsin D. Endocrinology 1982;

Yoshinari M , Taurog A. Lysosomal digestion of thyroglobulin: role of cathepsin D and thiol proteases. Endocrinology 1985;

36. Starling JR, Hopps BA. Inhibition of thyroid cathepsin D activity by pepstatin. Surg Res 1980; 28:8-13.

1 1 1:280-9. 35.

117:1621-31.

37. Dunn AD. Stimulation of thyroidal thiol endopeptidases by thyrotropin. Endocrinology 1984; 114:375-82.

38. Moscatelli D, Rifkin DB. Membrane and matrix localization of proteinases: a common theme in tumor cell invasion and angio- genesis. Biochim Biophys Acfa 1988; 94867-85.

39. Tandon AK, Clark GM, Chamness GC, Chirgwin JM, McGuire WL. Cathepsin D and prognosis in breast cancer. N Engl] Med

40. Mazzaferri EL, Young RL, Oertel JE, Kemmerer WT, Page CP. Papillary thyroid carcinoma: the impact of therapy in 576 patients. Medicine 1977; 56:171-96.

41. Tennvall J, Biorklund A, Moller T, Ranstam J, Akerman M. Prog- nostic factors of papillary, follicular and medullary carcinomas of the thyroid gland: retrospective multivariate analysis of 216 patients with a median follow-up of 11 years. Acta Radiol Oncol

1990; 322:297-302.

1985; 24:17-24.

estrogen receptors and cathepsin d in human thyroid tissue

Documents