Thyroid Cancer, Nonmedullary, 1

A number sign (#) is used with this entry because of evidence that susceptibility to nonmedullary thyroid cancer-1 (NMTC1) is conferred by heterozygous mutation in the thyroid transcription factor-1 gene (TITF1), also known as NK2 homeobox-1 (NKX2-1; 600635), on chromosome 14q13.

Nonmedullary thyroid cancer (NMTC) comprises thyroid cancers of follicular cell origin and accounts for more than 95% of all thyroid cancer cases. The remaining cancers originate from parafollicular cells (medullary thyroid cancer, MTC; 155240). NMTC is classified into 4 groups: papillary, follicular (188470), Hurthle cell (607464), and anaplastic. Approximately 5% of NMTC is hereditary, occurring as a component of a familial cancer syndrome (e.g., familial adenomatous polyposis, 175100; Carney complex, 160980) or as a primary feature (familial NMTC or FNMTC). Papillary thyroid cancer (PTC) is the most common histologic subtype of FNMTC, accounting for approximately 85% of cases (summary by Vriens et al., 2009).

PTC is characterized by distinctive nuclear alterations including pseudoinclusions, grooves, and chromatin clearing. PTCs smaller than 1 cm are referred to as papillary microcarcinomas. These tumors have been identified in up to 35% of individuals at autopsy, suggesting that they may be extremely common although rarely clinically relevant. PTC can also be multifocal but is typically slow-growing with a tendency to spread to lymph nodes and usually has an excellent prognosis (summary by Bonora et al., 2010).

Genetic Heterogeneity of Susceptibility to Nonmedullary Thyroid Cancer

Other susceptibilities to nonmedullary thyroid cancer include NMTC2 (188470), caused by mutation in the SRGAP1 gene (606523); NMTC3 (606240), mapped to chromosome 2q21; NMTC4 (616534), caused by mutation in the FOXE1 gene (602617); and NMTC5 (616535), caused by mutation in the HABP2 gene (603924).

A susceptibility locus for familial nonmedullary thyroid carcinoma with or without cell oxyphilia (TCO; 603386) has been mapped to chromosome 19p.

NMTC1

Ngan et al. (2009) identified 4 of 20 unrelated patients with multinodular goiter (MNG)/papillary thyroid carcinoma (PTC) who had an ala339-to-val (A339V) mutation in the TITF1 gene (600635.0012). Among the 4 patients with the A339V mutation, 2 women had first-degree relatives who also carried the mutation; all those relatives had had a history of MNG before diagnosis of PTC. One of the family members carrying this mutation developed metastatic colon cancer. One patient developed MNG at age 26 years; at age 37, she noticed a gradual increase in size of the goiter, and was found to have stage II disease. A second patient with this mutation was diagnosed at age 21 years with benign MNG and underwent right hemithyroidectomy. At the age of 48, she developed a left-sided thyroid swelling, which showed a follicular lesion necessitating total thyroidectomy. A 1.2-cm PTC was identified. The index case of family 2 developed benign MNG at age 34 years and PTC at age 46. The tumor contained a BRAF V600E mutation (164757.0001). The patient's mother was diagnosed with MNG in her twenties and PTC in her thirties; at age 71, she was diagnosed with colorectal carcinoma.

Familial Nonmedullary Thyroid Cancer

Lote et al. (1980) identified 2 kindreds with 7 and 4 cases of papillary carcinoma in otherwise healthy, nonirradiated subjects. All grew up in 1 of 2 small fishing villages in northern Norway. The familial cases showed an earlier mean age at diagnosis (37.6 years) than did sporadic cases from the same region (52.8 years). Multiple endocrine adenomatosis, Gardner syndrome (175100), and arrhenoblastoma (see 138800) were excluded.

Phade et al. (1981) described 3 affected sibs, of normal parents, with discovery of cancer at ages 12, 7, and 20 years. The authors found one other report of familial papillary carcinoma without polyposis coli, in a father and daughter, aged 40 and 12, respectively, at discovery (Lacour et al., 1973). The young age at occurrence and frequent bilateral involvement are characteristic of hereditary cancers.

Stoffer et al. (1985, 1986) presented evidence for the existence of a familial form of papillary carcinoma of the thyroid, possibly inherited as an autosomal dominant. Four parents of patients with familial PACT had colon cancer and 5 other family members died of intraabdominal malignancy that was not further defined. Perkel et al. (1988) presented evidence suggesting a familial susceptibility factor in radiation-induced thyroid neoplasms.

Grossman et al. (1995) identified 13 families with 30 individuals affected by familial nonmedullary thyroid cancer, which they abbreviated FNMTC. In 14 of these affected individuals whom they personally treated, 13 had multifocal tumors, and 6 of these were bilateral. The incidence of lymph node metastasis was 57%, as was the incidence of local invasion. Recurrences occurred in 7 patients during follow-up. The histologic diagnosis was papillary thyroid carcinoma in 13 of the 14 patients; in 1 patient it was Hurthle cell carcinoma.

Takami et al. (1996) identified 34 families in Japan with 72 individuals affected by nonmedullary thyroid cancer: 17 men and 55 women. Pathologic diagnosis was papillary carcinoma in 64 patients, follicular carcinoma in 6, and anaplastic carcinoma in 2. From the findings in their study they concluded that familial nonmedullary thyroid cancer behaves more aggressively than sporadic nonmedullary thyroid cancer.

Canzian et al. (1998) noted that families with multiple cases of nonmedullary thyroid cancer had been reported by Lote et al. (1980) and Burgess et al. (1997). FNMTC may represent 3 to 7% of all thyroid tumors. The tumors are usually multifocal, recur more frequently, and show an earlier age at onset than in sporadic cases. These characteristics are well exemplified by familial adenomatous polyposis-associated thyroid carcinoma, which, in addition, has been found to be a distinct morphologic entity, rather than the papillary carcinoma that it had previously been believed to be (Harach et al., 1994).

Vascular endothelial growth factor (VEGF; 192240) is a potent stimulator of endothelial cell proliferation that has been implicated in tumor growth of thyroid carcinomas. Using the VEGF immunohistochemistry staining score, Klein et al. (2001) correlated the level of VEGF expression with the metastatic spread of 19 cases of thyroid papillary carcinoma. The mean score +/- standard deviation was 5.74 +/- 2.59 for all carcinomas. The mean score for metastatic papillary carcinoma was 8.25 +/- 1.13 vs 3.91 +/- 1.5 for nonmetastatic papillary cancers (P less than .001). By discriminant analysis, they found a threshold value of 6.0, with a sensitivity of 100% and a specificity of 87.5%. The authors concluded that VEGF immunostaining score is a helpful marker for metastasis spread in differentiated thyroid cancers. They proposed that a value of 6 or more should be considered as high risk for metastasis threat, prompting the physician to institute a tight follow-up of the patient.

Baudin et al. (2003) studied the positive predictive value of serum thyroglobulin (TG; 188450) level after thyroid hormone withdrawal, measured during the first 6 to 12 months of follow-up in 256 consecutive differentiated thyroid cancer patients. They confirmed that (131)I-total body scan (TBS) has a limited interest for the follow-up of thyroid cancer patients. They concluded that follow-up should rely on serum TG level and prognostic parameters; however, initial serum TG may be produced by thyroid tissues of various significance, an increase at 2 consecutive determinations indicating disease progression and a decrease being related to late effects of therapy. The best positive predictive value is obtained by the slope of serum TG levels.

Serum TG assays are sometimes unsatisfactory for monitoring thyroid cancer because interference caused by anti-TG antibodies may reduce the sensitivity of the tests during thyroid hormone therapy. Savagner et al. (2002) developed a complementary method using real-time quantitative RT-PCR based on the amplification of TG mRNA. Two different pairs of primers were used for the determination of the frequency of 1 of the variants of the alternative splicing of TG mRNA. The frequency of this variant was as high in 40 patients as in 30 controls, accounting for about 33% of the total TG mRNA. Using appropriate primers, the authors observed that TG mRNA values in controls varied according to the volume of thyroid tissue and the TSH concentration. The TG mRNA values allowed the definition of a positive cutoff point at 1 pg/microg total RNA. This cutoff point, tested on the group of patients treated for thyroid cancer, produced fewer false negative results than those obtained with serum TG assays.

Wagner et al. (2005) tested the preoperative sensitivity of RT-PCR for TG and TSHR mRNA to detect thyroid cancer. TSHR and TG mRNA transcripts were detected by RT-PCR assays previously determined to be specific for cancer cells. There was 100% concordance between TSHR and TG mRNA RT-PCR results. The authors concluded that the molecular detection of circulating thyroid cancer cells by RT-PCR for TSHR/TG mRNA complements fine-needle aspiration cytology in the preoperative differentiation of benign from malignant thyroid disease, and that their combined use may save unnecessary surgeries. Wagner et al. (2005) suggested that this method shows promise for detecting follicular carcinoma, which is often missed by fine-needle aspiration cytology.

Carlomagno et al. (2002) showed that a pyrazolopyrimidine known as PP1 is a potent inhibitor of the RET kinase. Carlomagno et al. (2003) showed that another compound of the same class, known as PP2, blocks the enzymatic activity of the isolated RET kinase and RET/PTC1 oncoprotein at IC50 (inhibitory concentration-50; the amount of drug required to reduce activity in cell culture by 50%) in the nanomolar range. PP2 blocked in vivo phosphorylation and signaling of the RET/PTC1 oncoprotein. PP2 prevented serum-independent growth of RET/PTC1-transformed NIH 3T3 fibroblasts and of TPC1 and FB2, 2 human papillary thyroid carcinoma cell lines that carry spontaneous RET/PTC1 rearrangements. Growth in type I collagen (see 120150) gels efficiently reflects invasive growth of malignant cells. PP2 blocked invasion of type I collagen matrix by TPC1 cells. The authors concluded that pyrazolopyrimidines hold promise for the treatment of human cancers sustaining oncogenic activation of RET.

Fortunati et al. (2004) evaluated the action of valproic acid, a potent anticonvulsant reported to inhibit histone deacetylase, on cultured thyroid cancer cells. NPA (papillary or poorly differentiated) and ARO (anaplastic) cells were treated with increasing valproic acid concentrations. Expression of mRNA and cell localization pattern for the sodium-iodide symporter (NIS; 601843), as well as iodine-125 uptake, were evaluated before and after treatment. Valproic acid induced NIS gene expression, NIS membrane localization, and iodide accumulation in NPA cells, and it was effective at clinically safe doses in the therapeutic range. In ARO cells, only induction of NIS mRNA was observed, and was not followed by any change in iodide uptake. The authors concluded that valproic acid is effective at restoring the ability of NPA cells to accumulate iodide.

Oncogenic Rearrangements in Papillary Thyroid Carcinoma

Pierotti et al. (1996) indicated that oncogenic rearrangements of the RET gene are found in about 35% of cases of papillary thyroid carcinoma; rearrangements involving the NTRK1 gene are involved in about 15% of cases. The RET and NTRK1 genes encode membrane receptor-like proteins with tyrosine kinase activity. Their expression is strictly regulated and confined to subsets of neural crest-derived cells. The oncogenic rearrangements cause deletion of the N-terminal domain and fusion of the remaining tyrosine kinase domain of the receptor genes with the 5-prime end of different unrelated genes, designated activating genes. Since all the activating genes are ubiquitously expressed and also contain a dimerization domain, each RET and NTRK1 rearrangement produces chimeric mRNAs and proteins in the thyroid cells in which rearrangements occur. Moreover, the fusion products express an intrinsic and constitutive tyrosine kinase activity.

Among 329 thyroid lesions analyzed cytogenetically, Frau et al. (2008) identified 9 nodules with trisomy 17 as the only chromosomal change. All 9 cases were noninvasive, exhibited follicular growth pattern, and showed PTC-specific nuclear changes focally defined within the nodule. Frau et al. (2008) concluded that isolated trisomy 17 is associated with focal papillary carcinoma changes in follicular-patterned thyroid nodules and may be a marker for this poorly characterized subset of thyroid lesions.

Oncogenic Rearrangements in Follicular Thyroid Carcinoma

Kroll et al. (2000) demonstrated that the translocation t(2;3)(q13;p25), involving the fusion of the genes PAX8 (167415) and PPARG (601487), is a frequent event in human thyroid follicular carcinoma. Dwight et al. (2003) detected the PAX8/PPAR-gamma rearrangement by RT-PCR, FISH, and/or Western analysis in 10 of 34 (29%) follicular thyroid carcinomas and in 1 of 20 (5%) atypical follicular thyroid adenomas, but not in any of the 20 follicular thyroid adenomas or 13 anaplastic thyroid carcinomas studied. In addition, 7 of 87 thyroid tumors exhibited involvement of PPAR-gamma alone. The authors concluded that PAX8/PPAR-gamma occurs frequently in follicular thyroid carcinomas, and that the presence of this rearrangement may be highly suggestive of a malignant tumor.

RET Fusion Genes

In the case of the chimeric gene PTC1, RET is fused to the H4 gene (CCDC6; 601985), which, like RET, is located on chromosome 10 and becomes fused with RET through an intrachromosomal rearrangement. The chimeric gene PTC3 results from a structural rearrangement between RET with the ELE1 gene (NCOA4; 601984) on chromosome 10, and the chimeric gene PTC2 is generated through fusion of RET with the PRKAR1A gene (188830) on chromosome 17.

Corvi et al. (2000) identified a rearrangement involving the RET tyrosine kinase domain and the 5-prime portion of PCM1 (600299) on chromosome 8p22-p21. Immunohistochemistry using an antibody specific for the C-terminal portion of PCM1 showed that the protein level was drastically decreased and its subcellular localization altered in papillary thyroid tumor tissue with respect to normal thyroid.

By RT-PCR screening of PTCs of 2 patients exposed to radioactive fallout after the Chernobyl nuclear power plant disaster, followed by 5-prime RACE, Klugbauer et al. (1998) identified a novel RET rearrangement, PTC5, involving fusion of the RET tyrosine kinase domain to RFG5 (GOLGA5; 606918) on chromosome 14q.

Klugbauer and Rabes (1999) identified 2 novel types of RET rearrangements, which they termed PTC6 and PTC7. In PTC6, RET is fused to the N-terminal part of transcriptional intermediary factor-1-alpha (TIF1A; 603406) on chromosome 7q32-q34, and in PTC7, RET is fused to a C-terminal part of TIF1-gamma (TIF1G; 605769) on chromosome 1p13.

Herrmann et al. (1991) found clonal abnormalities on cytogenetic analysis in 9 out of 26 papillary thyroid cancers and 5 follicular thyroid cancers. In the former group, the abnormalities included loss of the Y chromosome, addition of an extra chromosome 5, or inversion in chromosome 10, inv(10)(q11.2q21.2). Using DNA probes specific for chromosomes 1, 3, 10, 16, and 17, they carried out RFLP analyses of 12 papillary cancers. No loss of heterozygosity (LOH) was observed for loci mapped to chromosome 10. Jenkins et al. (1990) likewise found the inv(10)(q11.2q21) with breakpoints where RET and another sequence of unknown function, D10S170 (H4; 601985), are located. Among 18 cases of papillary thyroid carcinoma, Pierotti et al. (1992) identified 5 with the identical abnormality. They reported the cytogenetic and molecular characterization of 4 of these tumors and demonstrated that the cytogenetic inversion provided the structural basis for the D10S170/RET fusion, leading to the generation of the chimeric transforming sequence which they referred to as RET/PTC. Santoro et al. (1992) found the activated form of the RET oncogene in 33 (19%) of 177 papillary carcinomas and in none of 109 thyroid tumors of other histotypes.

Bongarzone et al. (1994) examined tumors from a series of 52 patients with papillary thyroid carcinomas and identified 10 cases of RET fusion with the D10S170 locus (also known as H4) resulting in the generation of the RET/PTC1 oncogene, 2 cases with the gene encoding the regulatory subunit RI-alpha of protein kinase A (PRKAR1A; 188830), and 6 cases with a newly discovered gene they called ELE1 (601984) located on chromosome 10 and leading to the formation of the RET/PTC3 oncogene. The RET/PTC3 hybrid gene was expressed in all 6 cases and was associated with the synthesis of 2 constitutively phosphorylated isoforms of the oncoprotein (p75 and p80). The chromosome 10 localization of both RET and ELE1 and the detection, in all cases, of a sequence reciprocal to that generating the oncogenic rearrangements, strongly suggested that RET/PTC3 formation is a consequence of an intrachromosomal inversion of chromosome 10. The RET/PTC3 hybrid oncogene was observed in both sporadic and radiation-associated post-Chernobyl papillary thyroid carcinomas.

Bongarzone et al. (1997) examined the genomic regions containing the ELE1/RET breakpoints in 6 sporadic and 3 post-Chernobyl tumors in 2 papillary carcinomas of different origins. Notably, in all sporadic tumors and in 1 post-Chernobyl tumor, the ELE1/RET recombination corresponded with short sequences of homology (3 to 7 bp) between the 2 rearranging genes. In addition, they observed an interesting distribution of the post-Chernobyl breakpoints in the ELE1 break cluster region (bcr) located within an Alu element, or between 2 closely situated elements, and always in AT-rich regions.

NTRK1 Fusion Genes

In about 15% of cases of papillary thyroid carcinoma, the NTRK1 protooncogene (191315) is activated through fusion with neighboring genes TPM3 (191030) and TPR (189940) on chromosome 1q, and TFG (602498) on chromosome 3.

AKAP9/BRAF Fusion Gene

Ciampi et al. (2005) reported an AKAP9 (600409)-BRAF (164757) fusion that was preferentially found in radiation-induced papillary carcinomas developing after a short latency, whereas BRAF point mutations were absent in this group. Ciampi et al. (2005) concluded that in thyroid cancer, radiation activates components of the MAPK pathway primarily through chromosomal paracentric inversions, whereas in sporadic forms of the disease, effectors along the same pathway are activated predominantly by point mutations.

Lesueur et al. (1999) performed a linkage analysis on 56 informative kindreds collected through an international consortium on NMTC. Linkage analysis using both parametric and nonparametric methods excluded MNG1, TCO, and RET as major genes of susceptibility to NMTC and demonstrated that this trait is characterized by genetic heterogeneity.

In a genomewide association study of 192 Icelandic individuals with thyroid cancer and 37,196 controls, Gudmundsson et al. (2009) identified associations with SNPs on chromosomes 9q22.33 and 14q13.3, respectively. The findings were replicated in 2 cohorts of European descent (342 and 90 thyroid cancer cases, respectively). Overall, the strongest association signals were observed for rs965513 on 9q22.33 (see NMTC4, 616534) (odds ratio of 1.75; p = 1.7 x 10(-27)) and rs944289 on 14q13.3 (odds ratio of 1.37; p = 2.0 x 10(-9)). The gene nearest the 9q22.33 locus is thyroid transcription factor-2 (FOXE1; 602617) and thyroid transcription factor-1 (NKX2-1; 600635) is among the genes located at the 14q13.3 locus. Both variants contributed to an increased risk of both papillary and follicular thyroid cancer. Approximately 3.7% of individuals were homozygous for both variants, and their estimated risk of thyroid cancer was 5.7-fold greater than that of noncarriers. In large sample set from the general Icelandic population, both risk alleles were associated with low concentrations of TSH, and the 9q22.33 allele was associated with low concentration of T4 and high concentration of T3.

Jendrzejewski et al. (2012) found that rs944289 is located in a CEBP-alpha (CEBPA; 116897)/CEBP-beta (189965)-binding element in the 5-prime UTR of PTCSC3 (614821), a noncoding gene. They presented evidence suggesting that the risk allele of rs944289 decreases PTCSC3 promoter activation by reducing CEBP-alpha and CEBP-beta binding affinity compared with the nonrisk allele and thereby predisposes to papillary thyroid carcinoma.

Radiation-Related PTC

Takahashi et al. (2010) conducted a genomewide association study employing Belarusian patients with papillary thyroid cancer (PTC) aged 18 years or younger at the time of the Chernobyl accident and age-matched Belarusian control subjects. Two series of genome scans were performed using independent sample sets, and association with radiation-related PTC was evaluated. Metaanalysis combining the 2 studies identified 4 SNPs at chromosome 9q22.33 showing significant associations with the disease. The association was further reinforced by a validation analysis using one of these SNP markers, rs965513, with another set of samples. rs965513 is located 57 kb upstream to FOXE1 (602617), a thyroid-specific transcription factor with pivotal roles in thyroid morphogenesis and was reported as the strongest genetic risk marker of sporadic PTC in European populations. Of interest, no association was obtained between radiation-related PTC and rs944289 at 14q13.3, which showed the second strongest association with sporadic PTC in Europeans. The authors suggested that the complex pathway underlying the pathogenesis may be partly shared by the 2 etiologic forms of PTC, but their genetic components do not completely overlap each other, suggesting the presence of other unknown etiology-specific genetic determinants in radiation-related PTC.

The world's highest incidence of thyroid cancer has been reported among females in New Caledonia, a French overseas territory in the Pacific located between Australia and Fiji. Chua et al. (2000) investigated the prevalence and distribution of RET/PTC 1, 2, and 3 in papillary thyroid carcinoma from the New Caledonian population and compared the pattern with that of an Australian population. Fresh-frozen and paraffin-embedded papillary carcinomas from 27 New Caledonian and 20 Australian patients were examined for RET rearrangements by RT-PCR with primers flanking the chimeric region, followed by hybridization with radioactive probes. RET/PTC was present in 70% of the New Caledonian and in 85% of the Australian samples. Multiple rearrangements were detected and confirmed by sequencing in 19 cases, 4 of which had 3 types of rearrangements in the same tumor. The authors concluded that this study demonstrates a high prevalence of RET/PTC in New Caledonian and Australian papillary carcinoma. The findings of multiple RET/PTC in the same tumor suggested that some thyroid neoplasms may indeed by polyclonal.

Hrafnkelsson et al. (2001) studied the incidence of thyroid cancer in the relatives of Icelandic individuals in whom a diagnosis of nonmedullary thyroid cancer was made in the period 1955 to 1994. They identified 712 cases. The relative risk for thyroid cancer in all relatives was 3.83 for male relatives and 2.08 for female. The risk was highest in the male relatives of male probands (6.52) and lowest in the female relatives of female probands (2.02). For first-degree relatives the risk ratios were 4.10 for male relatives and 1.93 for female relatives.

Abubaker et al. (2008) studied the relationship of genetic alterations in the PIK3CA gene with various clinicopathologic characteristics of PTC in a Middle Eastern population. PIK3CA amplification was seen in 265 (53.1%) of 499 PTC cases analyzed, and PIK3CA gene mutations in 4 (1.9%) of 207 PTC. N2-RAS mutations were found in 16 (6%) of 265 PTC, and BRAF mutations in 153 (51.7%) of 296 PTC. NRAS mutations were associated with an early stage and lower incidence of extrathyroidal extension, whereas BRAF mutations were associated with metastasis and poor disease-free survival in PTCs. Abubaker et al. (2008) noted that the frequency of PIK3CA amplification was higher than that observed in Western and Asian populations, and remained higher after the amplification cutoff was raised to 10 or more.

RET/PTC Rearrangements

Sugg et al. (1998) examined the expression of RET/PTC-1, -2, and -3 in human thyroid microcarcinomas and clinically evident PTC to determine its role in early-stage versus developed PTC and to examine the diversity of RET/PTC in multifocal disease. Thirty-nine occult papillary thyroid microcarcinomas from 21 patients were analyzed. Of the 30 tumors (77%) positive for RET/PTC rearrangements, 12 were positive for RET/PTC1, 3 for RET/PTC2, 6 for RET/PTC3, and 9 for multiple RET/PTC oncogenes. In clinically evident tumors, 47% had RET/PTC rearrangements. Immunohistochemistry demonstrated close correlation with RT-PCR-derived findings. The authors concluded that RET/PTC expression is highly prevalent in microcarcinomas and occurs more frequently than in clinically evident PTC (P less than 0.005). Multifocal disease, identified in 17 of the 21 patients, exhibited identical RET/PTC rearrangements within multiple tumors in only 2 patients; the other 15 patients had diverse rearrangements in individual tumors. The authors inferred that RET/PTC oncogene rearrangements may play a role in early-stage papillary thyroid carcinogenesis, but seem to be less important in determining progression to clinically evident disease. In multifocal disease, the diversity of RET/PTC profiles, in the majority of cases, suggested to Sugg et al. (1998) that individual tumors arise independently in a background of genetic or environmental susceptibility.

By RT-PCR, Learoyd et al. (1998) analyzed the 3 main RET/PTC rearrangements and RET tyrosine kinase domain sequence expression in a prospective study of 50 adult PTCs. The genetic findings were correlated with the MACIS clinical prognostic score and with individual clinical parameters. Three of the patients had been exposed to radiation in childhood or adolescence. Four of the PTCs contained RET/PTC1, confirmed by sequencing, and none contained RET/PTC2 or RET/PTC3. The prevalence of RET rearrangements was 8% overall, but in the subgroup of 3 radiation-exposed patients it was 66.6%. Interestingly, RET tyrosine kinase domain mRNA was detectable in 70% of PTCs using RET exon 12/13 primers, and was detectable in 24% of PTCs using RET exon 15/17 primers. RT-PCR for calcitonin and RET extracellular domain, however, was negative. There was no association between the presence or absence of RET/PTC in any patient's tumor and clinical parameters. Learoyd et al. (1998) concluded that RET/PTC1 is the predominant rearrangement in PTCs from adults with a history of external irradiation in childhood.

Finn et al. (2003) assessed the prevalence of the common RET chimeric transcripts RET/PTC1 and RET/PTC3 in a group of sporadic PTCs and correlated them with tumor morphology. Thyroid follicular cells were laser capture microdissected from sections of 28 archival PTCs. Total RNA was extracted and analyzed for expression of glyceraldehyde 3-phosphate dehydrogenase (138400), RET/PTC1, and RET/PTC3 using TaqMan PCR. Ret/PTC rearrangements were detected in 60% of PTCs. Specifically, transcripts of RET/PTC1 and RET/PTC3 were detected in 43% and 18% of PTCs, respectively. Ret/PTC3 was detected in only follicular variant subtype (60%) and was not detected in classic PTC. One case of tall cell variant demonstrated chimeric expression of both RET/PTC1 and RET/PTC3 transcripts within the same tumor.

A sharp increase in the incidence of pediatric PTC was documented after the Chernobyl power plant explosion. An increased prevalence of rearrangements of the RET protooncogene (RET/PTC rearrangements) had been reported in Belarussian post-Chernobyl papillary carcinomas arising between 1990 and 1995. Thomas et al. (1999) analyzed 67 post-Chernobyl pediatric PTCs arising in 1995 to 1997 for RET/PTC activation; 28 were from Ukraine and 39 were from Belarus. The study, conducted by a combined immunohistochemistry and RT-PCR approach, demonstrated a high frequency (60.7% of the Ukrainian and 51.3% of the Belarussian cases) of RET/PTC activation. A strong correlation was observed between the solid-follicular subtype of PTC and the RET/PTC3 isoform: 19 of 24 (79%) RET/PTC-positive solid-follicular carcinomas harbored a RET/PTC3 rearrangement, whereas only 5 had a RET/PTC1 rearrangement. The authors concluded that these results support the concept that RET/PTC activation played a central role in the pathogenesis of PTCs in both Ukraine and Belarus after the Chernobyl accident.

Fenton et al. (2000) examined spontaneous PTC from 33 patients (23 females and 10 males) with a median age of 18 years (range, 6-21 years) and a median follow-up of 3.5 years (range, 0-13.4 years). RET/PTC mutations were identified in 15 tumors (45%), including 8 PTC1 (53%), 2 PTC2 (13%), 2 PTC3 (13%), and 3 (20%) combined PTC mutations (PTC1 and PTC2). This distribution is significantly different from that reported for children with radiation-induced PTC. There was no correlation between the presence or type of RET/PTC mutation and patient age, tumor size, focality, extent of disease at diagnosis, or recurrence. The authors concluded that RET/PTC mutations are (1) common in sporadic childhood PTC, (2) predominantly PTC1, (3) frequently multiple, and (4) of different distribution than that reported for children with radiation-induced PTC.

Elisei et al. (2001) evaluated the pattern of RET/PTC activation in thyroid tumors from different groups of patients (exposed or not exposed to radiation, children or adults, with benign or malignant tumors). They studied 154 patients, 65 with benign nodules and 89 with papillary thyroid cancer. In the last group, 25 were Belarus children exposed to the post-Chernobyl radioactive fallout, 17 were Italian adults exposed to external radiotherapy for benign diseases, and 47 were Italian subjects (25 children and 22 adults) with no history of radiation exposure. Among patients with benign thyroid nodules, 21 were Belarus subjects (18 children and 3 adults) exposed to the post-Chernobyl radioactive fallout, 8 were Italian adults exposed to external radiation on the head and neck, and 36 were Italian adults with naturally occurring benign nodules. The overall frequency of RET/PTC rearrangements in papillary thyroid cancer was 55%. The highest frequency was found in post-Chernobyl children and was significantly higher (P = 0.02) than that found in Italian children not exposed to radiation, but not significantly higher than that found in adults exposed to external radiation. No difference of RET/PTC rearrangements was found between samples from irradiated (external x-ray) or nonirradiated adult patients, as well as between children and adults with naturally occurring thyroid cancer. RET/PTC rearrangements were also found in 52.4% of post-Chernobyl benign nodules, in 37.5% of benign nodules exposed to external radiation and in 13.9% of naturally occurring nodules (P = 0.005, between benign post-Chernobyl nodules and naturally occurring nodules). The relative frequency of RET/PTC1 and RET/PTC3 in rearranged benign tumors showed no major difference. The authors concluded that the presence of RET/PTC rearrangements in thyroid tumors is not restricted to the malignant phenotype, is not higher in radiation-induced tumors compared with those naturally occurring, is not different after exposure to radioiodine or external radiation, and is not dependent on young age.

Mechler et al. (2001) reported 6 cases of familial PTC associated with lymphocytic thyroiditis in 2 unrelated families. PTC was diagnosed on classic nuclear and architectural criteria, and was bilateral in 5 cases. Architecture was equally distributed between typical PTC and its follicular variant. Lymphocytic thyroiditis was present in variable degrees, including, in 4 cases, oncocytic metaplasia. By use of RT-PCR, Mechler et al. (2001) demonstrated RET/PTC rearrangement in the carcinomatous areas of patients of both families: PTC1 in family 1, PTC3 in family 2, and a RET/PTC rearrangement in nonmalignant thyroid tissue with lymphocytic thyroiditis in family 2. The findings suggested that the molecular event at the origin of the PTCs was particular to each of the studied families, and confirmed that RET protooncogene activating rearrangement is an early event in the thyroid tumorigenic process and that it may occur in association with lymphocytic thyroiditis.

Zhu et al. (2006) analyzed 65 papillary carcinomas for RET1/PTC1 and RET/PTC3 using 5 different detection methods. The results suggested that broad variability in the reported prevalence of RET1/PTC arrangement is at least in part a result of the use of different detection methods and tumor genetic heterogeneity.

Germline Mutation in NKX2-1

Ngan et al. (2009) identified 4 of 20 unrelated patients with multinodular goiter (MNG)/papillary thyroid carcinoma (PTC) who had an ala339-to-val (A339V) mutation in the thyroid transcription factor-1 (TITF1) gene (NKX2-1; 600635.0012). Three of the 4 patients had more advanced tumors than did the remaining 16 patients. The mutation was not found among 349 healthy control subjects or among 284 PTC patients who had no history of MNG. Patients carrying the mutation had a higher incidence of perineural infiltration, but it was not statistically significant. Patients carrying the mutation were more likely than those without the mutation to have had previous thyroid surgery (50% vs 4.0%, p less than 0.001) and MNG (100% vs 5.3%, p less than 0.001).

Somatic Mutation in BRAF

Kimura et al. (2003) identified a val600-to-glu (V600E; 164757.0001) mutation in the BRAF gene in 28 (35.8%) of 78 cases of PTC; it was not found in any of the other types of differentiated follicular neoplasms arising from the same cell type (0 of 46). RET/PTC mutations and RAS (see 190020) mutations were each identified in 16.4% of PTCs, but there was no overlap in the 3 mutations. Kimura et al. (2003) concluded that thyroid cell transformation to papillary cancer takes place through constitutive activation of effectors along the RET/PTC-RAS-BRAF signaling pathway.

Namba et al. (2003) determined the frequency of BRAF mutations in thyroid cancer and their correlation with clinicopathologic parameters. The V600E mutation was found in 4 of 6 cell lines and 51 (24.6%) of 207 thyroid tumors. Examination of 126 patients with papillary thyroid cancer showed that BRAF mutation correlated significantly with distant metastasis (P = 0.033) and clinical stage (P = 0.049). The authors concluded that activating mutation of the BRAF gene could be a potentially useful marker of prognosis of patients with advanced thyroid cancers.

Xing et al. (2004) detected the V600E mutation in the BRAF gene in thyroid cytologic specimens from fine-needle aspiration biopsy (FNAB). Prospective analysis showed that 50% of the nodules that proved to be PTCs on surgical histopathology were correctly diagnosed by BRAF mutation analysis on FNAB specimens; there were no false-positive findings.

Other Somatic Mutation

In all of 6 examples of follicular thyroid carcinoma (FTC), Herrmann et al. (1991) found loss of heterozygosity (LOH) for RFLP markers on the short arm of chromosome 3. Such was not found in any of 3 follicular adenomas (FA) or 12 PTCs. Herrmann et al. (1991) suggested that a tumor suppressor gene on 3p is important for the development or progression of FTC.

Trovato et al. (1999) tested the hypothesis that both FTC and anaplastic thyroid cancer (ATC), but not PTC, could harbor LOH in segments of 7q encompassing the protooncogenes HGF (142409) and MET (164860). They screened 6 normal thyroids, 10 colloid nodules, 10 follicular hyperplasias, 10 oncocytic adenomas, 10 FAs, 10 FTCs, 6 ATCs, and 12 PTCs using 2 microsatellite markers for HGF and 2 for MET. LOH for all 4 markers was found in 100% of FTCs, 100% of ATCs, and (for only 1 or 2 markers) in 10 to 29% of FAs. The authors concluded that loss of genetic material explains why FTC and ATC, but not PTC, fail to express both HGF and MET.

Kitamura et al. (2001) carried out a genomewide allelotyping study of 66 follicular thyroid carcinomas using 39 microsatellite markers representing all nonacrocentric autosomal arms. The mean frequency of loss of heterozygosity was 9.2%, and the mean fractional allelic loss was 0.09. The most frequent allelic losses were detected in 7q (28%), 11p (28%), and 22q (41%). Frequent allelic losses of markers on chromosome 7q, 11p, and 22q suggested locations to examine for the presence of suppressor genes associated with the development of follicular thyroid carcinoma.

Nikiforova et al. (2003) identified a somatic mutation in the NRAS gene (Q61R; 164790.0002) in 70% (12) of follicular carcinomas and 55% (6) of follicular adenomas studied.

Garcia-Rostan et al. (2005) analyzed 13 thyroid cancer cell lines, 80 well-differentiated follicular (WDFTC) and papillary (WDPTC) thyroid carcinomas, and 70 anaplastic thyroid carcinomas (ATC) for activating PIK3CA (171834) mutations at exons 9 and 20. Nonsynonymous somatic mutations were found in 16 (23%) ATC cases, 2 (8%) WDFTC cases, and 1 (2%) WDPTC case. In 18 of 20 ATC cases showing coexisting differentiated carcinoma, mutations, when present, were restricted to the ATC component. Garcia-Rostan et al. (2005) concluded that mutant PIK3CA is likely to function as an oncogene in anaplastic thyroid carcinoma but less frequently in well-differentiated thyroid carcinomas.

Liu et al. (2008) explored a wide-range genetic basis for the involvement of genetic alterations in receptor tyrosine kinases (RTKs) and phosphatidylinositol 3-kinase (PI3K)/Akt and MAPK pathways in anaplastic thyroid cancer (ATC) and FTC. They found frequent copy gains of RTK genes including EGFR (131550) and VEGFR1 (165070), and PIK3CA and PIK3CB (602925) in the P13K/Akt pathway. RTK gene copy gains were preferentially associated with phosphorylation of Akt, suggesting their dominant role in activating the P13K/Akt pathway. Liu et al. (2008) concluded that genetic alterations in the RTKs and P13K/Akt and MAPK pathways are extremely prevalent in ATC and FTC, providing a strong genetic basis for an extensive role of these signaling pathways and the development of therapies targeting these pathways for ATC and FTC, particularly the former.

LOH of Imprinted Regions

Sarquis et al. (2006) investigated the hypothesis that in thyroid neoplasias loss of imprinted loci becomes enriched during oncogenesis. They studied thyroid tissue from 72 patients with thyroid neoplasias comprising 34 follicular thyroid carcinomas and 38 follicular adenomas. Overall LOH frequencies for the imprinted region (IR) markers were 26% for the adenomas and 38% for the carcinomas. In the nonimprinted regions (NIR), the overall LOH frequency was 23% and 26% for FAs and FTCs, respectively. The difference in LOH frequencies between IRs and NIRs was statistically significant only for the carcinomas (p = 0.001), although there was a similar trend for the atypical adenomas (p = 0.06). Sarquis et al. (2006) concluded that IRs are more prone to genomic instability in FTCs.

Weber et al. (2005) studied the frequency and mechanism of ARHI (605193) silencing in benign and malignant thyroid neoplasia. They demonstrated that underexpression of ARHI occurs principally in FTC (p = 0.0018), including its oncocytic variant (11 of 13), even at minimally invasive stage, but not classic PTCs (2 of 7) or follicular adenoma (FA) (3 of 14). FTC showed strong allelic imbalance with reduction in copy number/LOH in 69%, compared with less than 10% for FA. In combination with LOH data, bisulfite sequencing in a subset of samples revealed a symmetric methylation pattern for FA, likely representing 1 unmethylated allele and 1 presumptively imprinted allele, whereas FTC showed a virtually complete methylation pattern, representing LOH of the nonimprinted allele with only the hypermethylated allele remaining. Weber et al. (2005) showed that pharmacologic inhibition of histone deacetylation, but not demethylation, could reactivate ARHI expression in the FTC133 FTC cell line. Weber et al. (2005) concluded that silencing of the putative maternally imprinted tumor suppressor gene ARHI, primarily by large genomic deletion in conjunction with hypermethylation of the genomically imprinted allele, serves as a key early event in follicular thyroid carcinogenesis.