RET/Papillary Thyroid Cancer Rearrangement in Nonneoplastic ...

0021-972X/06/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 91(6):2414 –2423 Copyright © 2006 by The Endocrine Society doi: 10.1210/jc.2006-0240

RET/Papillary Thyroid Cancer Rearrangement in Nonneoplastic Thyrocytes: Follicular Cells of Hashimoto’s Thyroiditis Share Low-Level Recombination Events with a Subset of Papillary Carcinoma Kerry J. Rhoden,* Kristian Unger,* Giuliana Salvatore, Yesim Yilmaz, Volodymyr Vovk, Gennaro Chiappetta, Mazin B. Qumsiyeh, Jay L. Rothstein, Alfredo Fusco, Massimo Santoro, Horst Zitzelsberger, and Giovanni Tallini J. B. Pierce Laboratory (K.J.R.), School of Medicine Departments of Genetics (Y.Y., M.B.Q.) and Pathology (G.T.), Yale University, New Haven, Connecticut 06520; Institute of Molecular Radiobiology (K.U., V.V., H.Z.), Forschungszentrum fu¨r Umwelt und Gesundheit-National Research Center for Environment and Health, D-85764 Neuherberg, Germany; Department of Cellular and Molecular Pathology/Istituto di Endocrinologia ed Oncologia Sperimentale (G.S., A.F., M.S.), Consiglio Nazionale delle Ricerche, Universita di Napoli Federico II, 80131 Naples, Italy; Istituto Nazionale dei Tumori (G.C.), Fondazione Pascale, 80131 Naples, Italy; Kimmel Cancer Center, Department of Microbiology/Immunology and Otolaryngology-Head and Neck Surgery (J.L.R.), Thomas Jefferson University, Philadelphia, Pennsylvania 19107; and Naples Oncogenomic Center-CEINGE (A.F.), Biotecnologie Avanzate, 80145 Naples, Italy Context: RET/papillary thyroid cancer (PTC) is a marker for papillary thyroid carcinoma, but its specificity has been questioned because of the disputed identification of RET/PTC in Hashimoto’s thyroiditis (HT), oncocytic tumors, and other thyroid lesions. Objective: The objective of this study was to determine 1) whether RET/PTC occurs in nonneoplastic follicular cells of HT, and 2) its recombination rate in thyroid tumors. Design/Patients: Forty-three samples from 31 cases of HT were examined using interphase fluorescence in situ hybridization (FISH) with RET probes spanning the breakpoint region; real-time RT-PCR to quantify RET/PTC1, RET/PTC3, and c-RET transcripts; and RTPCR after laser capture microdissection to enrich samples for follicular cells. The results were compared with those similarly obtained in 34 papillary carcinomas, eight thyroid oncocytic tumors, and 21 normal thyroids.

signals were localized to rare nonneoplastic follicular cells; low-level RET/PTC was identified in 17% (five of 29) of thyroiditis cases by real-time RT-PCR and in an additional six of 11 real-time negative cases after increasing sensitivity with laser capture microdissection. Low RET/PTC1 levels were detected in 26% (nine of 34) of papillary carcinomas with an expression pattern and proportion of FISHpositive cells similar to those of the thyroiditis. Forty-seven percent (16 of 34) of papillary carcinomas and one oncocytic carcinoma expressed high RET/PTC1 mRNA levels. Conclusions: Low-level RET/PTC recombination occurs in nonneoplastic follicular cells in HT and in a subset of papillary thyroid carcinomas. RET/PTC expression variability should be taken into account for the molecular diagnosis of thyroid lesions. Overlapping molecular mechanisms may govern early stages of tumor development and inflammation in the thyroid. (J Clin Endocrinol Metab 91: 2414 –2423, 2006)

Results: Normal samples showed no RET rearrangement. Sixtyeight percent (15 of 22) of HT were positive by FISH; in all thyroiditis,

R

ET/PTC ONCOGENIC ACTIVATION is the result of chromosomal rearrangements that fuse the RETtyrosine kinase (TK) domain to the 5⬘-terminal region of heterologous genes. The rearrangements are a marker for papillary thyroid carcinoma (PTC) as their very name, RET/ PTC, implies. RET/PTC1 and RET/PTC3 are by far the most common types of RET rearrangement (1). Despite the proven association of RET/PTC with papillary

First Published Online April 4, 2006 * K.J.R. and K.U. have contributed equally to this paper. Abbreviations: Ct, Threshold cycle; FISH, fluorescence in situ hybridization; HT, Hashimoto’s thyroiditis; LCM, laser capture microdissection; NC, nuclear changes; PTC, papillary thyroid cancer; TK, tyrosine kinase. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

carcinoma, controversial studies have detected RET/PTC in greater than 90% of histologically benign Hashimoto’s thyroiditis (HT), suggesting the frequent occurrence of multiple occult neoplasms in the thyroiditis (2, 3). Moreover, the identification of RET/PTC in other common thyroid tumor histotypes such as oncocytic adenomas and carcinomas (4) and even in hyperplastic thyroid nodules (5, 6) seems to challenge the validity of RET/PTC as a tumor marker and its specificity for papillary carcinoma. It has been shown recently that the level of RET/PTC expression in papillary carcinoma is highly variable, suggesting that lesional distribution of RET/PTC may not be homogenous (7, 8). It is unknown whether variable RET/PTC expression levels may also explain the finding of RET/PTC transcripts in HT or other thyroid lesions. Therefore, we have systematically analyzed HT, papillary thyroid carcinoma, oncocytic tumors, and normal thyroid samples using three independent approaches. Expression

2414

Rhoden et al. • Low-Level RET/PTC in HT

J Clin Endocrinol Metab, June 2006, 91(6):2414 –2423

levels for RET exon 10 –11 and 12–13 transcripts adjacent to the rearrangement site on c-RET, RET/PTC1, and RET/PTC3 were measured by real-time RT-PCR using only high-quality RNA extracted from frozen samples. In selected cases of HT, laser capture microdissected material was further processed for highly sensitive RET expression analysis. RET rearranged DNA was directly visualized on histology sections by fluorescence in situ hybridization (FISH) using differentially labeled probes specific for the RET recombination site and laser scanning microscopy, which allows complete analysis of the nucleus throughout the thickness of the section. Patients and Methods Tissue samples and cell lines Thyroid samples were obtained from the Department of Pathology at Yale University School of Medicine. Excess lesional and/or perilesional tissue from surgically excised specimens was selected for the study. Patients underwent surgery for thyroid gland enlargement and/or for tumor resection. Tumors included 34 consecutive papillary carcinomas, eight oncocytic tumors (two carcinomas, six adenomas), and one medullary thyroid carcinoma. Forty-three samples were analyzed from different lesional areas of 31 HT (five associated with the papillary carcinomas mentioned above, 26 without associated thyroid carcinoma). Lesional HT areas with hyperplastic nodules and with severe inflammatory changes were selected. Twenty-one normal thyroid samples were included as controls. Each case was reviewed and representative sections of the frozen samples were cut and examined microscopically to confirm the adequacy of the sample and the diagnosis. Tumors were classified according to established criteria (9). The diagnosis of HT was determined histologically based on the presence of diffuse lymphoplasmacytic infiltration with germinal centers, parenchymal atrophy

2415

with oncocytic change, and variable amounts of stromal fibrosis. Cytological follicular cell alterations reminiscent of papillary carcinoma such as clearing, overlapping, and irregularity of the nuclear contour (PTC-NC, papillary carcinoma-type nuclear changes; Tables 1 and 2) were estimated and graded on a scale from 0 –3 in the HT samples analyzed (10). Fourteen HT samples were obtained from circumscribed, nonencapsulated hyperplastic nodules usually with poorly defined borders, but distinct from the atrophic appearing parenchymal background. The hyperplastic foci were composed of follicular cells with a variable (often prominent) proportion of oncocytic cells (9). These 14 samples from nodular HT (9) were analyzed as a distinct subset. None of them had the histological features of oncocytic adenoma, oncocytic carcinoma, or papillary carcinoma. None of the oncocytic tumors were associated with HT. HT was associated with 10 papillary carcinomas, including the five cases where both neoplastic and nonneoplastic samples were separately analyzed (see above). None of the normal samples exhibited any significant lymphocytic infiltrate. Handling of samples and clinical information proceeded in accordance with internal review boardapproved protocols. Cell lines used as positive controls were the TT medullary thyroid carcinoma cell line expressing c-RET (CRL-1803, American Type Culture Collection, Manassas, VA), the TPC1 cell line expressing the RET/PTC1 transcript (11), and the ARO undifferentiated thyroid carcinoma cell line transfected with a RET/PTC3 construct (7). The nontransfected ARO undifferentiated thyroid carcinoma cell line (10) was used as negative control.

Real-time quantitative RT-PCR Aliquots of fresh tissue were snap frozen and kept at ⫺80 C before RNA extraction for real-time RT-PCR. RNA extraction and quantitation were performed as formerly described (7). Primer and probe sequences used for this study were reported and validated in a previous report (7). RET exon 10 –11 and 12–13 and RET/PTC mRNA levels were normalized for ␤-actin mRNA and expressed as 2⫺⌬Ct, where ⌬Ct is the difference between the threshold cycles (Ct) for RET/PTC and ␤-actin

TABLE 1. Clinicopathological findings, RET expression, and FISH data in Hashimoto’s thyroiditisa Case ID

Inflammation (0 –3)b

PTC-NC (0 –3)c

RET/PTC (real-time RT-PCR)

H(P13)d

3

1

0

H(P23)d H(P24)d H(P34)d H11

2 3 3 3

0 2 0 3

0 0 NA 0

H12 H09 H13 H08 H07 H06 H01 H02 H10 H14 H05 H03

3 3 3 2 2 3 3 3 3 3 3 2

3 3 2 1 3 2 3 3 3 2 3 2

H04

3

3

0 0 0 0 0 0 0 0 0 0 0 sample a, low RET/PTC1; sample b, low RET/PTC1 Low RET/PTC1

cRET and RET/PTC (LCM-RT-PCR)

cRET⫺, RET/PTC1⫺, RET/PTC3⫹ NA NA NA Sample a, cRET⫹, RET/PTC3⫹, RET/PTC1⫺; sample b, cRET⫹, RET/PTC1⫺, RET/PTC3⫺ cRET⫺, RET/PTC1⫹, RET/PTC3⫺ RET/PTC1⫺, RET/PTC3⫺, cRET NA RET/PTC1⫺, RET/PTC3⫺, cRET NA RET/PTC1⫺, RET/PTC3⫺, cRET NA NA NA NA NA cRET⫺, RET/PTC1⫹, RET/PTC3⫺ NA NA NA

NA

Cell % with split RET signals

Cell % with ⫹10

sample a, 2.0; sample b, 1.4 0 NA 2.6 3.1

0 0 NA 0 0

3.3 3.9 5.6 6.2 0.9 5.8 NA NA NA NA NA sample a, 5.0; sample b, 4.0

0 0 0 0 0 0 NA NA NA NA NA sample a, 3.8; sample b, 3.0

6.0

0

ID, Identification number; NA, not available; PTC-NC, papillary carcinoma-type nuclear changes. a The results of multiple samples analyzed from the same case are indicated in alphabetical order. b Thyroiditis was graded on a scale from 0 –3 based on the number of infiltrating lymphocytes, plasma cells and other inflammatory cells, the number and size of germinal centers, and the extent of the damage to the thyroid parenchyma: 0, no significant infiltrate or parenchymal damage; 1, mild; 2, moderate; and 3, severe inflammation. All HT cases have grade 2 or 3 inflammation. c The extent of the PTC-NC in the follicular cells was graded on a scale from 0 –3: 0, rare PTC-NC; 1, few PTC-NC; 2, several PTC-NC; and 3, many PTC-NC. d Papillary carcinoma samples associated with HT were analyzed separately for cases P13, P16, P23, P24, and P34 (Table 3).

2416



TABLE 2. Clinicopathological findings, RET expression, and FISH data in nodular Hashimoto’s thyroiditisa Case ID

Size of hyperplastic nodule(s)

Inflammation (0–3)b

PTC-NC (0–3)c

HNOD10 HNOD12

1.2 cm sample a, 1.1 cm; sample b, 1.3 cm; sample c, background HT 3.0 cm sample a, 3.0 cm; sample b, background HT

3 3

2 sample a, 2; sample b, 1; sample c, 1 2 sample a, 3; sample b, 3

HNOD09 HNOD11

HNOD01 HNOD02 HNOD03 HNOD04 HNOD05 HNOD06 H(P16)d HNOD07 HNOD08

1.5 cm 2.8 cm sample a, 3.5 cm; sample b, background HT sample a, 1.0 cm; sample b, background HT sample a, 1.8 cm; sample b, background HT 2.2 cm sample a, 2.0 cm; sample b, background HT sample a, 3.0 cm; sample b, background HT sample a, 1.5 cm; sample b, background HT

3 3

2 3 2 3 2 3 3 3 2

2 2 sample a, 1; sample b, 1 sample a, 1; sample b, 1 sample a, 1; sample b, 1 2 sample a, 2; sample b, 1 sample a, 3; sample b, 1 sample a, 1; sample b, 1

cRET and RET/PTC (LCM-RT-PCR)


0 samples a and b, 0

cRET⫺, RET/PTC1⫺, RET/PTC3⫺ sample a, cRET⫹, RET/PTC1⫹, RET/PTC3⫹

0 0

NA sample a, cRET⫹, RET/PTC1⫹, RET/PTC3⫺; sample b, cRET⫺, RET/PTC1⫺, RET/PTC3⫺ NA NA NA

0 sample a, 4.5; sample b, 3.6; sample c, 0.8 6.4 sample a, 7.1; sample b, 3.6


0 0 samples a and b, 0

Cell % with ⫹10 0 0 0 0

NA NA sample a, 5.1

NA NA sample a, 11.3

samples a and b, 0

NA

sample a, 4.3

sample a, 5.4

samples a and b, 0

NA

NA

NA

NA sample a, low RET/PTC1; sample b, 0 sample a, low RET/PTC1; sample b, 0 sample a, low RET/PTC1

NA NA

8.4 sample a, 5.3; sample b, 1.8 sample a, 10.0; sample b, 0 sample a, 5.4; sample b, 3.7

0 samples a and b, 0

sample b, cRET⫺, RET/PTC1⫺, RET/PTC3⫺ NA

sample a, 3.1; sample b, 0 0

ID, Identification number; NA, not available; PTC-NC, papillary carcinoma-type nuclear changes. a The results of multiple samples analyzed from the same case are indicated in alphabetical order. b Thyroiditis was graded on a scale from 0 –3 based on the number of infiltrating lymphocytes, plasma cells and other inflammatory cells, the number and size of germinal centers, and the extent of the damage to the thyroid parenchyma: 0, no significant infiltrate or parenchymal damage; 1, mild; 2, moderate; and 3, severe inflammation. All HT cases have grade 2 or 3 inflammation. c The extent of the PTC-NC in the follicular cells was graded on a scale from 0 –3: 0, rare PTC-NC; 1, few PTC-NC; 2, several PTC-NC; and 3, many PTC-NC. d Papillary carcinoma samples associated with HT were analyzed separately for cases P13, P16, P23, P24, and P34 (Table 3). mRNA amplification. The efficiency of amplification was close to 1 for all genes (7). Assuming similar ␤-actin expression, normalized RET/ PTC mRNA levels were adjusted for the proportion of follicular vs. stromal and inflammatory cells estimated on hematoxylin and eosinstained frozen sections cut from the same sample used for RNA extraction. Quantitative RT-PCR data of 24 papillary carcinomas and 10 normal thyroid samples were previously reported (7).

Laser capture microdissection (LCM) RT-PCR LCM was performed using a PixCell I system (Arcturus Engineering, Mountain View, CA) as previously described (10). Approximately 1000 30-␮m shots were used to transfer on the thermoplastic film-coated cap cells obtained from each section of thyroid tissue. RNA extraction and RT-PCR with primers to detect c-RET, RET/PTC1, and RET/PTC3 were performed as described in earlier reports (11, 12). RT-PCR products were resolved on a 3% agarose gel, blotted, and hybridized with a probe covering the TK domain of RET (10).

Interphase FISH analysis Interphase FISH was performed on 10-␮m paraffin-embedded tissue sections (48 samples, including 25 of the 31 HT cases) following previously validated protocols (8) or on cytological touch preparations according to standard procedures from the same fresh snap-frozen tissue used for quantitative RT-PCR (24 samples) (13). DNA probe generation from three yeast artificial chromosome clones (313F4, 214H10, 344H4) covering the RET locus and probe hybridization were performed as detailed in an earlier report (8). Microscopic analyses were carried out using a confocal laser scanning microscope for histology sections (Zeiss LSM 510, Zeiss, Jena, Germany) and an epifluorescence microscope for cytological preparations (Olympus BX60, Olympus, Melville, CA). In either case, cell nuclei were scored for the presence of a split FISH signal, i.e. separated green (proximal to RET) and red (distal to RET) signals, in addition to an overlapping one. At least 100 nuclei with strong and well-delineated signals were scored. The proportion of thyroid follicular cells with split signals was assessed on the cytology imprints after adjustment for the number of inflammatory and stromal elements present on a hematoxylin and eosin-stained imprint

preparation performed in parallel with that analyzed by FISH. Interphase FISH analysis of histology sections enabled the direct selection of thyroid follicular cells that were scored on captured images as previously described (8). To determine a meaningful cutoff level indicating a significantly increased proportion of cells with split FISH signals compared with the baseline frequency, a series of normal thyroids were scored after careful review of the histology sections to exclude samples with any significant pathological alteration. These included five samples analyzed previously (8) and the 11 normal thyroids selected for interphase FISH analysis in this study. The mean value (x៮ ) and the sd of cells with split signals were determined (8). The resulting x៮ ⫾ sd values were 1.2 ⫾ 0.9. For calculation of the cutoff level, a value of 3 times x៮ , equivalent to 3.5% of aberrant cells with split RET signals, was accepted. No significant FISH score difference between cytology preparations and histology sections was observed considering all cases, as well as within each diagnostic category (normal, tumor, HT).

Statistical analysis The Mann-Whitney test, the Kruskal-Wallis test for nonparametric data with Dunn’s test for multiple comparisons, the Fisher’s exact test, and the ␹2 test for trend were used (7). Computations were performed with GraphPad Prism (GraphPad Software Inc., San Diego, CA).

Results

The clinicopathological features of the cases analyzed are summarized in Tables 1–3 and illustrated in Fig. 1. Five of 26 patients with HT and nine of the 43 patients with thyroid tumors were male. The median age of the patients with HT and with thyroid tumors was 43.5 and 44.0 yr, respectively. Real-time quantitative RT-PCR

RET exon 10 –11 and RET exon 12–13 mRNA expression. RET exon 10 –11 and 12–13 mRNA was detected in 12 of 14 normal



2417

TABLE 3. Clinicopathological findings, RET expression, and FISH data in papillary carcinoma and oncocytic tumors Case ID

P23b P33 P21 P32 P05 P07 P18 P19 P22 P06 P31 P17 P34b P20 P08 P24b P30 P28 P03 P26 P01 P15 P25 P09 P11 P02 P12 P04 P16b P14 P29 P27 P13b P10 O01 O02 O08 O03 O04 O05 O06 O07

Tissue sample

Tumor size (cm)

Background thyroiditis (0 –3)a



Cell % with biallelic split RET signals

Cell % with ⫹10 or ⫺10 (%)

PCcl PCcl PCcl PCcl PCfv PCfv PCfv PCcl PCcl PCcl PCcl PCfv PCcl PCcl PCfv PCcl PCfv PCcl PCds PCfv PCcl PCcl PCcl PCfv PCfv PCcl PCcl PCcl PCcl PCfv PCcl PCfv PCfv PCcl OA OC OA OA OA OA OA OC

1.8 3 1.5 1.7 1.4 3.5 3.3 1.2 3 2.4 1.1 1.2 2 3 3.8 4 3 5 5 4.5 5 4.5 4.2 2.5 3 2.2 3 4.8 1.2 1.9 2.5 4.9 4.5 3 2.5 3 1.3 3.5 1.2 2 2 3.8

2 0 0 3 2 0 0 3 0 0 0 0 3 0 0 3 0 0 2 0 1 0 0 0 1 0 2 1 3 0 1 0 3 1 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 Low RET/PTC1 Low RET/PTC1 Low RET/PTC1 Low RET/PTC1 Low RET/PTC1 Low RET/PTC1 Low RET/PTC1 Low RET/PTC1 Low RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1 High RET/PTC1, lowRET/PTC3 High RET/PTC1, lowRET/PTC3 High RET/PTC1, lowRET/PTC3 High RET/PTC1, lowRET/PTC3 High RET/PTC1, lowRET/PTC3 High RET/PTC1, lowRET/PTC3 0 0 0 0 0 0 0 High RET/PTC1, lowRET/PTC3

1.8 2.9 4 7.3 NA NA NA NA NA 10.8 NA 7.5 12.4 6.0 5.0 6.5 8.6 7.8 11.5 8 10.7 7.6 NA 6.7 15.2 4.7 NA 15.7 15.0 14.2 17.7 9.8 11.6 24.7 NA 0.8 4.5 5.7 NA NA NA 10.8

0 0 0 0 NA NA NA NA NA 3.6 NA 0 0 0 0 3.6 0 0 0 0 0 4.5 NA 0 0 0 NA 2.4 0 3.5 0 0 0 0 NA 0 0 0 NA NA NA 0

0 0 1.4 0 NA NA NA NA NA 3.6 (2.4) NA 0 0 0 0 1.2 (5) 0 0 13.0 0 10.7 NA 0 17.1 0 NA 5.6 (2.4) 0 0 0 2.0 0 1.4 NA 41.6 5.6 2.6 (11.7) NA NA NA 0

ID, Identification number; PCcl, papillary carcinoma classic; PCfv, papillary carcinoma follicular variant; PCds, papillary carcinoma diffuse sclerosing; OA, oncocytic adenoma; OC, oncocytic carcinoma; F, female; M, male; Y, yes; N, no; NA, not available; ⫹10, extra copy of chromosome 10; ⫺10, loss of one chromosome 10. a Background thyroiditis was graded on a scale from 0 –3 based on the number of lymphocytes, plasma cells, and other inflammatory cells; the number and size of germinal centers; and the extent of the damage to the thyroid parenchyma: 0, no significant lymphoid infiltrate or parenchymal damage; 1, mild; 2, moderate; and 3, severe inflammation. All cases with HT exhibited grade 2 or 3 inflammation. b Nontumor samples with HT were separately analyzed 关Table 1, cases H(P13), H(P16), H(P23), H(P24), and H(P34)兴.

thyroids (86%), 32 of 36 HT samples, corresponding to 27 of 29 HT cases (93%), in all papillary carcinomas, and in all oncocytic tumors (Fig. 2). Median RET exon 10 –11 and 12–13 mRNA levels were similar in normal thyroid, HT, and papillary carcinoma but were significantly lower in oncocytic tumors than in normal thyroids (P ⬍ 0.05). With the exception of the medullary carcinoma, median RET exon 10 –11 and 12–13 mRNA levels in normal and pathological samples were approximately 500-5000 times lower than in the c-RET control TT cell line. As a consequence of the RET/PTC rearrangement, the TK-encoding domain of RET (starting from exon 12) is placed under the transcriptional control of the strong promoters of the RET gene fusion partners; this results in the overexpres-

sion of the 3⬘ end of the RET gene (from exon 12 to the 3⬘-ter) with respect to the 5⬘ end (from exon 1–11). Therefore, RET exon 12–13 vs. exon 10 –11 mRNA expression levels were plotted (Fig. 2C), and the ratio of exon 12–13 to exon 10 –11 mRNA was calculated (Fig. 2D). In normal thyroid tissues, exon 12–13 was linearly related to exon 10 –11 expression (r2 ⫽ 0.99; slope ⫽ 0.99) (Fig. 2C), confirming that exon 12–13 mRNA expression in normal thyroid reflects wild-type cRET. Furthermore, the mean ratio of exon 12–13 to exon 10 –11 mRNA levels in normal thyroid, 1.07 ⫾ 0.24, was not significantly different from the theoretical value of 1 for perfectly balanced RET expression. A cutoff ratio of 3.2 (3 times the mean ratio in normal thyroid) was determined, below which RET expression was considered balanced and

2418



FIG. 1. Histological appearance of HT and papillary thyroid carcinoma with corresponding FISH images. Split FISH signals are highlighted and magnified in the square insets. A, Hematoxylin and eosin section of HT with a hyperplastic nodular area shown in the top right of the picture (case HNOD08, Tables 1 and 2) (magnification, ⫻100). B, FISH image of the hyperplastic area corresponding to the asterisk in A (magnification, ⫻400). C, FISH image of the thyroiditis outside the hyperplastic nodule in the area corresponding to the circle in A (magnification, ⫻400). D, Hematoxylin and eosin section of a papillary carcinoma (case P29, Tables 1 and 2) (magnification, ⫻200). E, FISH image of the same papillary carcinoma (magnification, ⫻400).

above which it was considered unbalanced (Fig. 2D). Sixteen papillary carcinomas and one oncocytic carcinoma of the eight oncocytic tumors (adenomas and carcinomas) analyzed had unbalanced exon 12–13 mRNA expression (Fig. 2, C and D), anticipating their positivity for RET/PTC rearrangement. RET/PTC1 and RET/PTC3 mRNA expression. RET/PTC1 mRNA was detected in 25 of 34 (74%) papillary carcinomas, six of 36 HT samples corresponding to five of 29 (17%) HT cases, the oncocytic carcinoma with unbalanced RET expression (one of eight or 13%), but in none of the 14 normal thyroids or in the medullary carcinoma (Fig. 3 and Tables 1–3). RET/PTC1 mRNA expression varied approximately 100,000-fold and differed between cases with balanced and unbalanced RET expression. Tumors with unbalanced RET expression had significantly higher RET/ PTC1 mRNA (median 2⫺⌬Ct ⫽ 5.7 ⫻ 10⫺2) than those with balanced RET expression (median 2⫺⌬Ct ⫽ 1.7 ⫻ 10⫺5, P ⬍

0.001). Thus, RET/PTC1-positive papillary carcinomas could be divided into two subsets: one with unbalanced RET expression and high RET/PTC1 level (2⫺⌬Ct ⬎ 10⫺3) corresponding to that expressed by the RET/PTC1-positive TPC1 cell line; another with balanced RET expression and low RET/PTC1 level (2⫺⌬Ct ⬍ 10⫺3), a level too low to be reflected in a detectable exon 10 –11/exon 12–13 RET expression imbalance (Fig. 3A). All HT samples positive for RET/PTC1 had balanced RET expression and low RET/PTC1 mRNA levels that, like those of the papillary carcinoma subset with low RET/PTC1, were too low to result in detectable exon 10 –11/exon 12–13 RET imbalance. RET/PTC1 mRNA levels in HT and in the low RET/PTC1 papillary carcinoma subset corresponded to a dilution of less than one RET/PTC1-positive TPC1 cell in 10,000 negative cells. RET/PTC1 amplicons were identified by direct gel visualization and ethidium bromide staining in



2419

FIG. 2. RET exon 12–13 and exon 10 –11 mRNA expression in normal and diseased thyroid. RET exon 10 –11 (A) and RET exon 12–13 (B) mRNA levels were measured in normal thyroid (N and 䉫), HT (H and ƒ), papillary carcinoma (P and E), oncocytic tumors (O and 䡺), medullary carcinoma (M and ‚), and the TT cell line (TT and ⫻) and expressed as 2⫺⌬Ct values with respect to ␤-actin mRNA. Numbers represent the proportion of samples with RET exon 10 –11 or exon 12–13 mRNA out of the total examined for each group. C, Plot of RET exon 12–13 vs. exon 10 –11 mRNA levels, with the linear regression fit of normal thyroid values (r2 ⫽ 0.99, slope ⫽ 0.99). D, Ratio of RET exon 12–13 to exon 10 –11 mRNA levels, with ratios greater than the cutoff value of 3.2 indicating unbalanced RET expression. Symbols represent individual samples, and horizontal bars represent the median for each group.

nine papillary carcinomas expressing high RET/PTC1 levels by real-time RT-PCR and in the RET/PTC1-positive oncocytic tumor. RET/PTC3 was detected in six of 34 papillary carcinomas (18%) and in the oncocytic carcinoma mentioned above (Fig. 3B). RET/PTC3 levels were similar to the RET/PTC1 levels in the set of papillary carcinomas with low RET/PTC1 expression (2⫺⌬Ct ⬍ 10⫺3). All RET/PTC3-positive cases had unbalanced RET expression and high levels of RET/PTC1 mRNA (2⫺⌬Ct ⬎ 10⫺3). LCM-RT-PCR

To assess the potential difficulty of identifying small numbers of RET/PTC-expressing cells within a large tissue specimen, we used LCM for better resolution. Selected specimens FIG. 3. RET/PTC1 and RET/PTC3 mRNA expression in diseased thyroid. RET/PTC1 mRNA (A) and RET/PTC3 mRNA (B) were measured in HT (H and ƒ), papillary carcinoma with balanced (Pbal and E), or unbalanced (Punb and F) exon 12–13 vs. exon 10–11 mRNA expression, oncocytic tumors (O and f), TPC1 cells (TPC1 and ⫻), and RET/PTC3-transfected Aro cells (Aro and *). RET/PTC mRNA levels are calculated as 2⫺⌬Ct values with respect to ␤-actin mRNA. For all disease categories, white and black symbols represent samples with balanced and unbalanced exon 12–13 vs. exon 10–11 mRNA expression, respectively. Numbers represent the proportion of samples with RET/PTC out of the total examined in each group, and horizontal bars represent the median for each group.

of HT with optimal tissue preservation, morphological follicular cell alterations, and no detectable RET/PTC transcripts by real-time RT-PCR were thus further processed for LCM-RT-PCR. The results are summarized in Tables 1 and 2. RET/PTC mRNA was detected in seven of 13 samples corresponding to six of 11 (55%) HT cases. RET/PTC was thus identified by real-time RT-PCR in 17% of HT (see above) and, after increasing detection sensitivity with LCM-RT-PCR, in 55% of those cases previously deemed negative. The proportion of HT with detectable RET/PTC combining both methods is therefore estimated at 62%. Analysis of separately microdissected samples from the same case demonstrated heterogeneity in the pattern of c-RET and RET/PTC expression (Tables 1 and 2).

2420



FIG. 4. FISH analysis of RET rearrangements in thyroid tissue. A, Proportion of cells with FISH split signal was quantified in normal thyroid (N and 䉫), nodular HT (nod), and HT without hyperplastic nodules (non, nonnodular thyroiditis) (ƒ), papillary carcinoma (P and E), and oncocytic tumors (O and 䡺); data for all samples analyzed; symbols represent individual samples. B, Samples with a proportion of FISH split signals greater than the cutoff of 3.5% illustrated as boxesand-whiskers; boxes extend from the 25th to 75th percentiles, the line in the middle represents the median, and whiskers reflect the range of values; numbers represent the proportion of cases with FISH split signals greater than the cutoff, as indicated by the dotted line.

Interphase FISH analysis

The proportion of follicular cells with split signals was significantly higher in HT (P ⬍ 0.05) and in papillary carcinoma (P ⬍ 0.001) compared with that identified in normal thyroids (Fig. 4A). Twenty-four of 26 papillary carcinomas (92%), 19 of 31 samples corresponding to 15 of 22 (68%) HT cases, and three of four (75%) oncocytic tumors had a proportion of cells with split RET signals above the cutoff level (Fig. 4B). Among samples with split FISH signals above the cutoff, papillary carcinomas had a higher median proportion of cells with split signals compared with HT (P ⬍ 0.001) (Fig. 4B). Although the median proportion of cells with split FISH signals was similar in HT samples with and without hyperplastic areas, the proportion of samples with FISH split signals above the cutoff was significantly higher for the hyperplastic group (P ⬍ 0.05) (Fig. 4B). In none of the HT were split signals localized to occult foci of papillary carcinoma. Extra chromosome 10 copies were detected in four of 22 (18%) HT cases, in one involving more than 10% of cells. All but one of the cases corresponded to hyperplastic areas within the thyroiditis (Tables 1 and 2). Chromosome 10 gains or losses were identified in 10 papillary carcinomas, involving more than 10% of the cells in three cases and in three oncocytic tumors (Table 3). A small proportion of cells with split FISH signals were identified in both chromosome 10 copies in five papillary carcinomas (Table 3). Papillary carcinomas with high (2⫺⌬Ct ⬎ 10⫺3) RET/PTC1 expression had a significantly higher median proportion of cells with split FISH signals than either HT (P ⬍ 0.05) or papillary carcinomas (P ⬍ 0.05) with low (2⫺⌬Ct ⬍ 10⫺3) RET/PTC1 mRNA (Fig. 5). Correlation of c-RET and RET/PTC rearrangements with clinicopathological features

In cases where multiple lesional areas of the thyroidectomy specimen were analyzed using real-time RT-PCR or FISH to detect RET/PTC, the highest scores were always associated with samples showing evidence of cellular proliferation exemplified by papillary carcinoma or hyperplastic foci within the thyroiditis (Tables 1–3). For instance, in case

P16, selected regions of the patient’s thyroid were sampled corresponding to papillary carcinoma, a hyperplastic nodule, and the background thyroiditis without follicular cell hyperplasia. The papillary carcinoma sample expressed high RET/PTC1 levels with low-level RET/PTC3 and had 15% cells with a split FISH signal. The hyperplastic nodule had low RET/PTC1 expression and 5.3% of cells with split FISH signals. The background thyroiditis sample had no detectable RET/PTC expression and split FISH signals below the cutoff level. Extensive follicular cell alterations reminiscent of papillary thyroid carcinoma (PTC-NC grade 2 and 3, Tables 1 and

FIG. 5. Relationship between RET rearrangement detected by interphase FISH and RET/PTC1 expression in thyroid tissue. The proportion of cells with FISH split signals was quantified in normal thyroid (䉫), HT (ƒ), papillary carcinoma (E), and oncocytic tumors (䡺) expressing 0, less than 10⫺3, or more than 10⫺3 RET/PTC1 mRNA relative to ␤-actin; symbols represent individual samples; horizontal bars represent the median for each group; the dotted line represents the 3.5% FISH cutoff.


2) were present in 12 of 15 (80%) HT cases positive by FISH, whereas the majority of HT with RET/PTC detectable by either real-time RT-PCR or after LCM also showed extensive cytological alterations reminiscent of papillary thyroid carcinoma and severe thyroiditis (Tables 1 and 2). Despite the trend, however, the correlation between cytological alterations and RET/PTC did not reach statistical significance. Among papillary carcinomas, high RET/PTC1 expression was statistically associated with large tumor size (P ⬍ 0.05) (Table 3). The median age of patients was 35 for those with papillary carcinoma expressing high RET/PTC1 and 44 for those with papillary carcinomas with no detectable or low RET/PTC1 expression; however, the difference did not reach statistical significance. Background thyroiditis (both nonspecific and HT) was present in nine of 16 (56%) papillary carcinomas with high RET/PTC1 but in six of 18 (33%) cases that were either negative or had low RET/PTC1 (Table 3). Discussion RET/PTC rearrangement in HT

Our data demonstrate RET recombination in follicular cells of HT, although very few of the cells harbor the rearrangement. RET rearrangement was detected in up to 68% of HT depending on the method used. We can confidently exclude the existence of microscopic foci of papillary carcinoma as the possible source of the rearrangement (3) because lesional tissue was directly analyzed by interphase FISH and by RT-PCR with mRNA obtained from laser-microdissected cells and because of the careful review of the samples before molecular analysis. The number of follicular cells with split FISH signals above background was always limited. The low number of FISH-positive cells correlated with the need for highly sensitive RT-PCR after laser microdissection to detect RET/PTC mRNA and with the balanced exon 10 –11/12–13 RET expression. It was also consistent with the low RET/PTC mRNA levels that were always equivalent to less than one RET/PTC1-positive TPC1 cell in 10,000 negative ones. These data have interesting implications. On one hand, the occurrence of low-level RET/PTC rearrangement (the identification of which is critically dependent on any combination of factors that affect sensitivity) likely accounts for the conflicting results among studies that have used different approaches to detect aberrant transcripts in HT (2, 3, 14). On the other hand, the demonstration of RET/PTC in nonneoplastic HT follicular cells is intriguing given the inflammatory nature of the process and the established oncogenic RET/PTC potential. Several observations have associated HT to thyroid tumors, including the occurrence in the thyroiditis of proliferative nodules, of cytological alterations such as nuclear clearing and irregularities similar to those required for the diagnosis of full blown papillary cancer, and a likely increase in the risk for papillary carcinoma (9). Follicular cells in HT can be karyotypically abnormal (15), an observation supported by the detection of low-level chromosome 10 trisomy in four of our cases, two of which had detectable RET/PTC1 by real-time RT-PCR. Allelic DNA loss and expression of molecular papillary carcinoma markers have also been reported (16 –18). Additional evidence suggests that RET/PTC


2421

may provide a link between HT and papillary carcinoma. In fact, patients exposed to the Chernobyl radioactive fallout develop not only RET/PTC-driven papillary cancer but also chronic autoimmune thyroiditis (19); moreover, transgenic mice expressing RET/PTC show chronic thyroiditis in addition to papillary carcinoma (20). Our demonstration of RET/PTC in nonneoplastic thyrocytes further supports the link among RET/PTC, HT, and papillary carcinoma. However, the consistent detection of low expression levels and the restriction of the rearrangement to few follicular cells indicate that RET/PTC does not necessarily anticipate the development of papillary carcinoma in patients with thyroiditis. Incidentally, Braf-activating mutations have not been identified in eight samples of lesional HT tissue microdissected and then analyzed by direct sequencing of amplified genomic DNA (data not shown). Two hypotheses may explain the association between HT and RET/PTC. According to one, the very inflammatory process favors the occurrence of the rearrangement. The propensity of thyrocytes to undergo RET recombination has been explained by the peculiar arrangement of chromatin that juxtaposes RET and its fusion partners in interphase nuclei. It is possible that free radical production, cytokine secretion, cellular proliferation, and other events related to the inflammation trigger the occurrence of the rearrangement in follicular cells predisposed to it by an unstable chromatin conformation (21). Similar mechanisms may underlie the occurrence of low-level RET rearrangement as a secondary phenomenon in papillary carcinoma subsets (see below). According to the second hypothesis, the occurrence of RET/ PTC may directly influence the inflammation of HT. In fact, RET/PTC protein expression produces a strong inflammatory response in experimental models (22) and activates numerous inflammatory mediators and molecules within thyroid epithelial cells (23–26). In this respect, recent observations that RET/PTC-transformed thyroid cells can modify their microenvironment through chemokine- and cytokinemediated autocrine/paracrine loops that promote autonomous cell proliferation in neighboring nonneoplastic thyrocytes is very intriguing (23). RET/PTC rearrangement in thyroid tumors

As far as tumors are concerned, two distinct expression patterns could be identified. One was characterized by very low RET/PTC1 mRNA levels, the other by high RET/PTC1 mRNA levels with or without low-level RET/PTC3 coexpression. RET/PTC1 mRNA levels in the first group of tumors were so low that they did not result in a detectable imbalance between RET exon 10 –11 and 12–13 expression and, like the FISH results, were similar to those of HT. Somatic mutations in cancer have been called driver when they are positively selected and causally related to tumor development and passenger when not directly implicated with tumor growth (27). RET/PTC can hardly be considered a driver mutation in the tumors with minimal levels of rearranged RET. Rather, and similar to the above discussion for HT, these rearrangements may reflect RET instability in thyroid follicular cells (21) and point to the existence of secondary cell subclones.

2422


Although it may be difficult to establish when a mutational event becomes relevant for tumor pathogenesis (27), our data demonstrate a clear demarcation between tumors with low and high RET/PTC rearrangement levels. RET rearrangement in small tumor cell subclones is likely to be of no consequence in determining biological behavior but must be taken into account for molecular diagnosis. Highly sensitive nonquantitative methods to detect RET/PTC should be avoided. The existence of papillary carcinomas with low RET/PTC rearrangement level provides indeed a reasonable explanation for the inconsistency of RET/PTC detection reported in the literature using methods with different sensitivity (for a review of RET/PTC prevalence rates, see Ref. 7). In the second group of tumors with high levels of rearrangement, RET/PTC is likely to be the driver alteration with an impact on the biological and clinical features. All cases with high RET/PTC1 expression were indeed malignant tumors, papillary carcinomas, and one oncocytic carcinoma. The driving role of RET/PTC is further supported by the lack of overlap between RET/PTC and BRAF or RAS mutations, the other two major oncogenic events in papillary carcinoma (28). Papillary carcinomas with high RET/PTC1 had in our study a lower median age at diagnosis, which correlates with previous observations (29). The finding of high-level RET/ PTC1 expression in one oncocytic carcinoma confirms previous reports indicating that RET rearrangement may have a pathogenetic role in at least a subset of these neoplasms (4). The coexpression of low-level RET/PTC3 in some of the tumors with high RET/PTC1 expression provides further evidence for the molecular heterogeneity of RET rearrangement. Although the relative proportion of the RET/PTC1 and RET/PTC3 rearranged forms has never been quantified before, the coexistence of both in the same tumor has been previously reported (30 –33), and the occurrence of more than two different RET rearrangement types within a single lesion is certainly consistent with multiple RET/PTC rearranged clones in at least some papillary carcinomas (30). In our cases, RET/PTC3 was only expressed at low levels and always in tumors with high RET/PTC1 expression, consistent with RET/PTC3 rearrangement as a secondary phenomenon in a micropopulation of tumor cells. Interestingly, some of the papillary carcinomas had low-level chromosome 10 aneuploidy and a few cells with split FISH signals in both copies of chromosome 10. From a technical point of view, although FISH scores in the tumors with high RET/PTC levels were significantly elevated compared with those in samples with little RET/PTC expression, they were still much lower than expected. Our FISH scores are very similar to those reported in a recent study of RET/PTC rearrangements in radiation-associated papillary carcinomas (8). Heterogeneous clustering of neoplastic cell carrying RET rearrangement is a possible explanation for the lower than expected FISH scores (8). However, comparison of FISH with quantitative real-time RT-PCR measurements indicates that in the high RET/PTC1 expression group of tumors, the rearrangement ought to be present in the large majority of the neoplastic cells. In fact, the existence of minor cell subclones is difficult to reconcile with the consistent detection of unbalanced RET expression and high RET/PTC1 levels, comparable with those of the RET/


PTC1-positive TPC1 cells. It must therefore be assumed that FISH underestimation of RET rearrangement and/or genetic heterogeneity within the tumor are responsible for the discrepancy between RET expression and FISH data (34). In summary, this study demonstrates that RET/PTC rearrangement is present in nonneoplastic HT follicular cells; therefore, RET/PTC detection per se should not be equated with a diagnosis of papillary carcinoma. It suggests that overlapping molecular mechanisms may govern early stages of tumor development and inflammation in the thyroid. It indicates that in some papillary carcinomas, RET/PTC represents a passenger mutation occurring in a minority of tumor cells. Quantitative methods to detect RET/PTC should be applied whenever possible (7), not only for the molecular diagnosis of thyroid lesions but also to stratify patients that could potentially benefit from novel therapeutic approaches based on recently discovered RET kinase inhibitors (1). Acknowledgments We thank Chaline Johnson and Elke Konha¨user for skillful technical assistance and Antonia Conti for help with the illustrations of the manuscript. Received February 2, 2006. Accepted March 24, 2006. Address all correspondence and requests for reprints to: Giovanni Tallini, M.D., Ospedale Bellaria, Via Altura 3, 40139 Bologna, Italy. E-mail: [email protected]. Present address for K.J.R. and G.T.: Laboratory of Medical Genetics, Department of Internal Medicine (K.J.R.), Department of Pathology (G.T.), University of Bologna School of Medicine, 40139 Bologna, Italy. This work was supported by the Department of Pathology, Yale University School of Medicine and in part by the Associazione Italiana per la Ricerca sul Cancro Regional Grant Reference Code 1145 (to G.T.).

References 1. Santoro M, Carlomagno F, Melillo RM, Fusco A 2004 Dysfunction of the RET receptor in human cancer. Cell Mol Life Sci 61:2954 –2964 2. Sheils OM, O’Eary JJ, Uhlmann V, Lattich K, Sweeney EC 2000 ret/PTC-1 activation in Hashimoto thyroiditis. Int J Surg Pathol 8:185–189 3. Wirtschafter A, Schmidt R, Rosen D, Kundu N, Santoro M, Fusco A, Multhaupt H, Atkins J, Rosen M, Keane W, Rothstein JL 1997 Expression of the RET/PTC fusion gene as a marker for papillary carcinoma in Hashimoto’s thyroiditis. Laryngoscope 107:95–100 4. Cheung CC, Ezzat S, Ramyar L, Freeman JL, Asa SL 2000 Molecular basis of Hurthle cell papillary thyroid carcinoma. J Clin Endocrinol Metab 85:878 – 882 5. Elisei R, Romei C, Vorontsova T, Cosci B, Veremeychik V, Kuchinskaya E, Basolo F, Demidchik EP, Miccoli P, Pinchera A, Pacini F 2001 RET/PTC rearrangements in thyroid nodules: studies in irradiated and not irradiated, malignant and benign thyroid lesions in children and adults. J Clin Endocrinol Metab 86:3211–3216 6. Ishizaka Y, Kobayashi S, Ushijima T, Hirohashi S, Sugimura T, Nagao M 1991 Detection of retTPC/PTC transcripts in thyroid adenomas and adenomatous goiter by an RT-PCR method. Oncogene 6:1667–1672 7. Rhoden KJ, Johnson C, Brandao G, Howe JG, Smith BR, Tallini G 2004 Real-time quantitative RT-PCR identifies distinct c-RET, RET/PTC1 and RET/ PTC3 expression patterns in papillary thyroid carcinoma. Lab Invest 84:1557– 1570 8. Unger K, Zitzelsberger H, Salvatore G, Santoro M, Bogdanova T, Braselmann H, Kastner P, Zurnadzhy L, Tronko N, Hutzler P, Thomas G 2004 Heterogeneity in the distribution of RET/PTC rearrangements within individual post-Chernobyl papillary thyroid carcinomas. J Clin Endocrinol Metab 89:4272– 4279 9. Rosai J, Ackerman LV 2004 Surgical pathology. XI ed. New York: Mosby; 520 –523 10. Fusco A, Chiappetta G, Hui P, Garcia-Rostan G, Golden L, Kinder BK, Dillon DA, Giuliano A, Cirafici AM, Santoro M, Rosai J, Tallini G 2002 Assessment of RET/PTC oncogene activation and clonality in thyroid nodules with incomplete morphological evidence of papillary carcinoma: a search for the early precursors of papillary cancer. Am J Pathol 160:2157–2167 11. Ishizaka Y, Ushijima T, Sugimura T, Nagao M 1990 cDNA cloning and


12.

13.

14. 15. 16.

17.

18. 19.

20.

21. 22. 23.

24.

characterization of ret activated in a human papillary thyroid carcinoma cell line. Biochem Biophys Res Commun 168:402– 408 Santoro M, Papotti M, Chiappetta G, Garcia-Rostan G, Volante M, Johnson C, Camp RL, Pentimalli F, Monaco C, Herrero A, Carcangiu ML, Fusco A, Tallini G 2002 RET activation and clinicopathologic features in poorly differentiated thyroid tumors. J Clin Endocrinol Metab 87:370 –379 Zitzelsberger H, Szucs S, Weier HU, Lehmann L, Braselmann H, Enders S, Schilling A, Breul J, Hofler H, Bauchinger M 1994 Numerical abnormalities of chromosome 7 in human prostate cancer detected by fluorescence in situ hybridization (FISH) on paraffin-embedded tissue sections with centromerespecific DNA probes. J Pathol 172:325–335 Nikiforova MN, Caudill CM, Biddinger P, Nikiforov YE 2002 Prevalence of RET/PTC rearrangements in Hashimoto’s thyroiditis and papillary thyroid carcinomas. Int J Surg Pathol 10:15–22 Vanni R, Marras-Virdis S, Gerosa C, Lai ML, Tallini G 2000 Cytogenetics of thyroid nodules in Hashimoto thyroiditis. Cancer Genet Cytogenet 120:87– 88 Gasbarri A, Sciacchitano S, Marasco A, Papotti M, Di Napoli A, Marzullo A, Yushkov P, Ruco L, Bartolazzi A 2004 Detection and molecular characterisation of thyroid cancer precursor lesions in a specific subset of Hashimoto’s thyroiditis. Br J Cancer 91:1096 –1104 Hunt JL, Baloch ZW, Barnes L, Swalsky PA, Trusky CL, Sesatomi E, Finkelstein S, LiVolsi VA 2002 Loss of heterozygosity mutations of tumor suppressor genes in cytologically atypical areas in chronic lymphocytic thyroiditis. Endocr Pathol 13:321–330 Prasad ML, Huang Y, Pellegata NS, de la Chapelle A, Kloos RT 2004 Hashimoto’s thyroiditis with papillary thyroid carcinoma (PTC)-like nuclear alterations express molecular markers of PTC. Histopathology 45:39 – 46 Pacini F, Vorontsova T, Molinaro E, Kuchinskaya E, Agate L, Shavrova E, Astachova L, Chiovato L, Pinchera A 1998 Prevalence of thyroid autoantibodies in children and adolescents from Belarus exposed to the Chernobyl radioactive fallout. Lancet 352:763–766 Santoro M, Chiappetta G, Cerrato A, Salvatore D, Zhang L, Manzo G, Picone A, Portella G, Santelli G, Vecchio G, Fusco A 1996 Development of thyroid papillary carcinomas secondary to tissue-specific expression of the RET/PTC1 oncogene in transgenic mice. Oncogene 12:1821–1826 Gandhi M, Medvedovic M, Stringer JR, Nikiforov YE 2006 Interphase chromosome folding determines spatial proximity of genes participating in carcinogenic RET/PTC rearrangements. Oncogene 25:2360 –2366 Powell Jr DJ, Eisenlohr LC, Rothstein JL 2003 A thyroid tumor-specific antigen formed by the fusion of two self proteins. J Immunol 170:861– 869 Melillo RM, Castellone MD, Guarino V, De Falco V, Cirafici AM, Salvatore G, Caiazzo F, Basolo F, Giannini R, Kruhoffer M, Orntoft T, Fusco A, Santoro M 2005 The RET/PTC-RAS-BRAF linear signaling cascade mediates the motile and mitogenic phenotype of thyroid cancer cells. J Clin Invest 115:1068 –1081 Puxeddu E, Knauf JA, Sartor MA, Mitsutake N, Smith EP, Medvedovic M, Tomlinson CR, Moretti S, Fagin JA 2005 RET/PTC-induced gene expression


25. 26. 27.

28.

29.

30.

31. 32. 33.

34.

2423

in thyroid PCCL3 cells reveals early activation of genes involved in regulation of the immune response. Endocr Relat Cancer 12:319 –334 Russell JP, Engiles JB, Rothstein JL 2004 Proinflammatory mediators and genetic background in oncogene mediated tumor progression. J Immunol 172:4059 – 4067 Shinohara S, Rothstein JL 2004 Interleukin 24 is induced by the RET/PTC3 oncoprotein and is an autocrine growth factor for epithelial cells. Oncogene 23:7571–7579 Davies H, Hunter C, Smith R, Stephens P, Greenman C, Bignell G, Teague J, Butler A, Edkins S, Stevens C, Parker A, O’Meara S, Avis T, Barthorpe S, Brackenbury L, Buck G, Clements J, Cole J, Dicks E, Edwards K, Forbes S, Gorton M, Gray K, Halliday K, Harrison R, Hills K, Hinton J, Jones D, Kosmidou V, Laman R, Lugg R, Menzies A, Perry J, Petty R, Raine K, Shepherd R, Small A, Solomon H, Stephens Y, Tofts C, Varian J, Webb A, West S, Widaa S, Yates A, Brasseur F, Cooper CS, Flanagan AM, Green A, Knowles M, Leung SY, Looijenga LH, Malkowicz B, Pierotti MA, Teh BT, Yuen ST, Lakhani SR, Easton DF, Weber BL, Goldstraw P, Nicholson AG, Wooster R, Stratton MR, Futreal PA 2005 Somatic mutations of the protein kinase gene family in human lung cancer. Cancer Res 65:7591–7595 Kimura ET, Nikiforova MN, Zhu Z, Knauf JA, Nikiforov YE, Fagin JA 2003 High prevalence of BRAF mutations in thyroid cancer: genetic evidence for constitutive activation of the RET/PTC-RAS-BRAF signaling pathway in papillary thyroid carcinoma. Cancer Res 63:1454 –1457 Bongarzone I, Vigneri P, Mariani L, Collini P, Pilotti S, Pierotti MA 1998 RET/NTRK1 rearrangements in thyroid gland tumors of the papillary carcinoma family: correlation with clinicopathological features. Clin Cancer Res 4:223–228 Chua EL, Wu WM, Tran KT, McCarthy SW, Lauer CS, Dubourdieu D, Packham N, O’Brien CJ, Turtle JR, Dong Q 2000 Prevalence and distribution of ret/ptc 1, 2, and 3 in papillary thyroid carcinoma in New Caledonia and Australia. J Clin Endocrinol Metab 85:2733–2739 Finn SP, Smyth P, O’Leary J, Sweeney EC, Sheils O 2003 Ret/PTC chimeric transcripts in an Irish cohort of sporadic papillary thyroid carcinoma. J Clin Endocrinol Metab 88:938 –941 Sugg SL, Ezzat S, Rosen IB, Freeman JL, Asa SL 1998 Distinct multiple RET/PTC gene rearrangements in multifocal papillary thyroid neoplasia. J Clin Endocrinol Metab 83:4116 – 4122 Tallini G, Santoro M, Helie M, Carlomagno F, Salvatore G, Chiappetta G, Carcangiu ML, Fusco A 1998 RET/PTC oncogene activation defines a subset of papillary thyroid carcinomas lacking evidence of progression to poorly differentiated or undifferentiated tumor phenotypes. Clin Cancer Res 4:287– 294 Fagin JA 2004 Challenging dogma in thyroid cancer molecular genetics: role of RET/PTC and BRAF in tumor initiation. J Clin Endocrinol Metab 89:4264 – 4266

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.