Purpose: To determine whether microsatellite instability (MSI) in particular loci has clinicopathological significance in thyroid cancer.

Experimental Design: Seventy-six cases of surgically resected thyroid cancer were screened for MSI at nine microsatellites: THRA1, TSHR, D2S123, D11S912, D2S115, D2S399, p53, RET, or BAT-26. Multivariate analysis was performed to test for links between MSI and the clinical parameters of gender, age, histology, stage, nodal involvement, and prognosis.

Results: THRA1, residing in the thyroid hormone receptor α gene, displayed the highest levels of MSI at 36.5%. MSI in TSHR, located within the thyroid-stimulating hormone receptor gene, was found to be linked to cancer in the elderly (>70 years of age) and with high-grade (N 3, 4) nodal involvement. In follicular cancer, MSI in D2S123 occurred at a frequency of 100% (7/7) with no (0%) occurrence of MSI at the nearby D2S115, D2S399, or BAT-26 loci. Regarding prognosis, patients with MSI-positive cancer showed better long-term survival. BAT-26, which is an important marker in colorectal cancer, displayed the lowest frequency of MSI in our panel of thyroid tumors.

Conclusion: Whereas patients with MSI-positive cancer showed better long-term survival, as is the case for colorectal cancer, our finding of the low frequency of MSI in BAT-26 suggests that the biochemical defects governing the spectrum of MSI in thyroid and colorectal cancer are different. MSI in THRA1, TSHR, and D2S123 appears to be an integral part of thyroid carcinogenesis, as evidenced by the high frequency of MSI and significant correlation to clinical data.

Human thyroid tumors derive from either epithelial follicular cells or from parafollicular C cells (1). Tumors of follicular cell origin range from benign adenomas through papillary and follicular cancers to undifferentiated anaplastic carcinoma. Follicular adenomas and carcinomas often show mutations in one of the three Ras genes (2) and occasionally in the PTEN tumor suppressor gene (3). PTC3 often show characteristic gene rearrangements, which give rise to the formation of several types of RET/PTC chimeric genes (4). Such genomic instability occurs in ∼50% of all of the papillary cancers and involves the juxtaposition of the 3′-tyrosine kinase domain of the RET proto-oncogene with the 5′ domain of a ubiquitously expressed gene, resulting in the constitutive activation of RET/PTC proteins (5, 6). Anaplastic carcinoma is associated frequently with mutations of the p53 tumor suppressor gene (7). Point mutations of the RET proto-oncogene are a common feature of medullary thyroid carcinoma, a malignant tumor derived from parafollicular C cells. Despite these beginnings, a conclusive relationship between specific gene alterations and particular forms of thyroid cancer has not yet been established (8).

MSI is a distinct form of genomic instability linked to defects in DNA mismatch repair (9). MSI is characterized by alterations in the length of simple homopolymeric sequences (i.e., microsatellites) that are ubiquitous in the genome. The presence of MSI in tumor tissue has been associated with unique clinical and pathological characteristics (10, 11, 12, 13). While most notably associated with hereditary nonpolyposis colon cancer, MSI has been found to be involved in the genesis of many types of cancer (14, 15). Although MSI was not initially observed in thyroid cancer (16), subsequent studies have documented MSI not only in thyroid cancer but also in benign goiters and follicular adenomas of the thyroid (17, 18, 19). These more recent studies support the notion that MSI plays an important role in thyroid cancer and disease.

Studies on genomic instability in cancer have revealed that MSI occurs much more frequently in certain loci than in others and that the spectrum of MSI varies with cancer type (20). In colorectal cancer, where the majority of past work has focused, a set of five microsatellites (BAT26, BAT25, D2S123, D5S346, and D17S250) has been selected by an international consortium of scientists to serve as a reference panel for assessment of MSI status (20). MSI in BAT-26 appears most strongly linked to the replication error phenotype in colorectal cancers and cell lines (21). The biological significance of the reference panel is still unclear outside of the observation that if MSI is observed in members of the reference panel then it is highly likely that MSI will be observed in a variety of alternative microsatellite loci. Overall, the reference panel of five microsatellites was chosen by consensus vote rather than asserting some defining criteria. As the link between MSI and cancer becomes more clear selection criteria will undoubtedly also be brought into better focus. The character of MSI in noncolonic tumors was reviewed by Boland et al.(20), and it was concluded that the reference panel of five microsatellites used to assess MSI in colorectal cancer is not suitable to serve as a reference panel for other types of cancer. Consequently, there is a need to find microsatellites appropriate for assessing MSI status in noncolonic tumors. In this report we present data on particular microsatellites that appear well suited for assessing MSI in thyroid cancer.

We chose to analyze MSI at locations with an association to thyroid cancer, focusing on microsatellites close to the RET, p53, hMSH2, TSHR, and THRA1 genes, and a microsatellite on chromosome 11. To date, mutations of the RET gene have the strongest link to thyroid cancer (8), whereas loss of heterozygosity affecting the p53 gene is often observed in anaplastic thyroid carcinoma (22). Although loss of heterozygosity in p53 has not been frequently observed in thyroid cancers of other histological types (5, 23, 24), the importance of p53 in carcinogenesis overall made an assessment of MSI near p53 prudent. Given the importance of defective DNA mismatch repair in the MSI phenotype, four microsatellites on chromosome 2 near hMSH2 were tested, namely, BAT-26, which is located in an intron of hMSH2, and D2S123, D2S115, and D2S399. We also chose to analyze TSHR, located within the thyroid stimulating hormone receptor gene, and THRA1, which resides in the thyroid hormone receptor α gene (25, 26). We included one additional microsatellite in our panel, namely D11S912, chosen because chromosome 11 carries a putative tumor suppressor gene important in thyroid tumorigenicity (27, 28).

Patients Profile and Clinical Data.

Seventy-six normal tumor-matched samples were collected at Fukushima Medical University Hospital from 1993 to 1999. Patient clinical data are shown in Table 1. There were 65 female and 11 male patients. Histopathological classification and staging are according to Union Internationale Contre le Cancer criteria (29). The average of survival follow-up term was 166.8 weeks. All of the cases were sporadic primary thyroid carcinoma except one multiple endocrine neoplasia type 2A case. In one portion of the study dealing with MSI at D2S123 in follicular carcinoma, 45 cases of benign follicular adenoma (7 male, average age 47.2 years, and 38 female, average age 50.9 years) were analyzed.

Sample Collection and DNA Extraction.

After surgical resection, tumor and normal control tissue was classified as to histology and stage by a medical pathologist then immediately frozen to −80°C. To obtain purified DNA the frozen samples were homogenized, treated with proteinase K, and the DNA isolated by phenol-chloroform extraction (30).

Microsatellite Analysis.

PCR-based microsatellite analysis was performed at the eight (CA)n dinucleotide repeat microsatellite markers, D2S115, D2S123, D2S399, RET, D11S912, TSHR, THRA1, and p53, and the one (A)26 mononucleotide microsatellite marker BAT-26. PCR primers were designed using sequence information from GenBank4 and Primer3 for primer selection5 except D11S912 and p53. For these two loci, we used PCR primers that were reported previously (28, 31). PCR primer sequences are presented in Table 2. The THRA1 locus exists in an exon, but the other loci exist in intronic regions. PCR amplifications were performed in a 10-μl reaction mixture containing 100 ng of DNA, 10 mm Tris-HCl (pH 8.4), 50 mm KCl, 100 μm deoxynucleotide triphosphates, 1.5 mm MgCl, 10 pmol of forward and reverse primers, 0.01% gelatin, 0.25 units of Taq DNA polymerase (Perkin-Elmer, Foster City, CA), and 1 mCi of [α-32P]dCTP (EASYTIDES; NEN Life Science Products Inc., Boston, MA). PCR was performed using an initial denaturing step at 94°C for 3 min, then steps of 94°C for 40 s, 54–60°C for 40 s, 72°C for 40 s, repeated for 30 cycles, followed by a final extension step of 72°C for 5 min. PCR amplification was performed on a PTC-100 PCR machine (MJ Research, Inc., Watertown, MA). For gel electrophoresis the PCR amplification products were first mixed with a formamide loading dye heat denatured and then loaded on 6% SequaGel (National Diagnostic, Atlanta, GA) containing Tris-borate EDTA buffer and 6 m urea. Electrophoresis on a standard vertical DNA sequencing gel apparatus was done at 1300 V for 2–3 h. After electrophoresis, gels were dried and used for autoradiography with Fuji HR-H film (Fuji Photo Film Co., Ltd., Tokyo, Japan) for 12–48 h.

Assessment of MSI.

To assess MSI, we compared the band pattern produced after gel electrophoresis of paired PCR reactions in which patient matched normal and tumor DNA was amplified. If the normal and tumor PCR amplification products displayed different electrophoretic mobility we scored the case as positive for MSI. If either of the PCR amplifications failed, we omitted the case from the judgement of MSI. Microdissection of tumor samples was not performed; consequently, the MSI frequencies we report are lower than or equal to MSI frequencies in absolutely pure tumor tissue. MSI was independently judged in a blinded manner by the authors.

Statistical Analysis.

To assess the statistical significance of relationships between MSI and clinical parameters, Student’s t test, χ2 test, and McNemar’s test were used. For survival rate, the Kaplan-Meier survival curve method was used with statistical significance by log-rank test. P < 0.05 was considered as statistically significant. All of the tests were performed with Statview version 5.0 software (SAS Institute Inc., San Francisco, CA). Multivariate analysis was performed, and only links supported by a P < 0.05 have been presented.

Frequency of MSI.

Representative gel electrophoresis band patterns of MSI positive and negative cases are shown in Fig. 1. At the eight (CA)n dinucleotide microsatellites tested, the following frequencies of MSI were observed: THRA1 (27/74, 36.5%); D2S123 (24/74, 32.4%); D11S912 (21/76, 27.6%); D2S115 (17/76, 22.4%); RET (12/64, 18.8%); TSHR (14/76, 18.4%); p53 (13/76, 17.1%); and D2S399 (10/70, 14.3%). The lowest frequency of MSI was observed at the mononucleotide (A)26 microsatellite BAT-26 (5/60, 8.3%), which is located in an intron of the DNA mismatch repair gene hMSH2.

The THRA1 (CA)18 microsatellite, residing in exon 9 of the thyroid hormone receptor α gene (25), showed the highest frequency of MSI in our study. The frequency of MSI in RET, TSHR, p53, D2S399, and BAT-26 was calculated to be significantly lower than that measured in THRA1 based on McNemar’s test with 95% confidence (see Table 3). The frequency of MSI in the D2S123, D11S912, and D2S115 markers was not significantly different from that measured in THRA1 based McNemar’s test within a 95% confidence interval. As a group, THRA1, D2S123, D11S912, and D2S115 showed on average MSI in 29.7% of the thyroid cancer tissues tested.

Relationship between MSI at One Locus and Average MSI.

To determine whether MSI status in one locus was indicative of an elevated overall frequency of MSI, we compared the frequency of MSI at each locus to the average MSI for our panel (Table 4). For example, in the 27 tumors displaying MSI at THRA1, the average MSI for our panel was 38.2% (see Table 4). In the 47 tumors that did not display MSI at THRA1, the average MSI for our panel was 16.1%. Therefore, if a tumor displayed MSI in THRA1, then MSI was more likely to be observed in other microsatellites in our panel compared with those tumors that did not exhibit MSI in THRA1. D2S123, D11S912, TSHR, p53, D2S115, and D2S399 also showed this trait (see Table 4). Interestingly, BAT-26 and RET did not display this characteristic. Tumors displaying MSI in BAT-26 or RET were not significantly different from the tumors not displaying MSI in BAT-26 or RET in regard to the frequency of MSI at the other microsatellites tested in this study.

Relationship between MSI and Clinicopathological Factors.

To clarify the relationship between MSI and various standard clinical parameters, we compared MSI status at each microsatellite with the following clinical parameters: gender; age; TNM classification; stage; histological type; recurrence; and nodal involvement (see Table 5). Significant correlation between MSI in certain microsatellite markers and histological type, patient age, nodal involvement, and tumor stage were observed. As to histology, all seven of the follicular carcinomas displayed MSI in the D2S123 locus, whereas in nonfollicular tumors MSI in D2S123 was observed at a frequency of 25.3% (17 of 67). The link between MSI in D2S123 and follicular carcinoma was calculated to be statistically significant (P = 0.0003). To test if the link between MSI in D2S123 extended to follicular adenoma, we screened for MSI at D2S123 in 45 cases of benign follicular adenoma (7 male, average age 47.2 years, and 38 female, average age 50.9 years). We found 7% (3/45) of the benign follicular adenoma cases displayed MSI at D2S123. A χ2 test as to the significance of the difference in MSI frequency yielded a P of 0.0001, which strongly supports the notion that MSI in D2S123 is linked solely to the follicular cancer phenotype.

Regarding patient age, in tumors from patients >70 years of age, MSI in TSHR was observed at a frequency of 50% (4 of 8), whereas for patients <70 years of age, the percentage of MSI in TSHR was 14.7% (10 of 68). The correlation between MSI in TSHR and patient age was significant (P < 0.05, χ2 test). In regard to nodal involvement, the N 3, 4 group of tumors showed a higher frequency of MSI in TSHR with 5 of 13 cases or 38.5% versus 9 of 63 cases or 14.3% in the N 0, 1, 2 group. The correlation between MSI in TSHR and nodal involvement was calculated to be statistically significant (P < 0.05, χ2 test).

Prognostic Value of MSI in Thyroid Carcinoma.

Survival data for 75 of the 76 thyroid cancer cases used in this study is known. To determine whether MSI was linked to survival the 75 cases were divided into two groups: a MSI-negative group with no MSI at any microsatellite tested and a MSI-positive group with MSI found in at least one microsatellite. For these two groups we compared survival (see Fig. 2). The MSI-negative group showed a much poorer prognosis than the MSI-positive group (P = 0.007, log-rank test). In the MSI-positive group the survival rate did not depend on whether additional loci displayed MSI (data were not shown).

The BAT-26 locus showed the lowest frequency of MSI in our panel and had the lowest predictive value as to the occurrence of MSI in other loci. For colorectal cancer, MSI in BAT-26 has been found to be a reliable predictor as to the occurrence of MSI in other loci (21) and has strong prognostic value (12). The significance of MSI in the BAT-26, located in intron 5 of the hMSH2 gene, has been taken as evidence of the importance of defective hMSH2-mediated mismatch repair in MSI-positive colorectal cancers (13, 21). This does not appear to be the case in thyroid cancer.

Our data on BAT-26 suggests that the biochemical defects giving rise to MSI in thyroid cancer are different from those responsible for MSI in colorectal cancer. Whereas the biochemistry governing the spectrum of MSI in cancer is unknown, our data indicate that BAT-26 is not a primary target of MSI in thyroid cancer. Interestingly, whereas thyroid and colorectal cancer appear quite different in regard to BAT-26, our observation that patients with MSI-positive tumors had significantly better prognosis parallels what has been found in colorectal cancer. Although exactly why survival is better in the case of MSI-positive tumors is still unclear, it does seem that a MSI-positive phenotype carries with it genetic alterations that manifest a less aggressive phenotype (12, 14). For thyroid cancer, MSI events outside of BAT-26 would appear to govern prognosis. The lower than expected frequencies and weak prognostic value of MSI in RET and p53 is taken as evidence that these loci are also not primary MSI targets in thyroid cancer and that other microsatellite markers should probably be used to evaluate MSI in thyroid cancer.

The THRA1 locus displayed the highest frequency of MSI (36.5%). The THRA1 gene is a homologue of v-erbA, which was one of the first examples of a dominant-negative oncogene (32). Immunohistochemical analysis of cancer cell lines (33) has substantiated that MSI in the THRA1 microsatellite, residing in exon 9 of the gene, affects THRA1 protein expression (data not shown). Whereas no clinicopathological relation with THRA1 was found, MSI in THRA1 was indicative of MSI in other loci with significant relation to clinical data. It is conceivable that MSI in THRA1 is linked to a clinical parameter we did not test or that MSI in this locus is associated with an obscure phenotype not easily characterized in a clinical setting. The same may be true for D2S115 and D11S912, because high MSI frequencies were also observed in these loci, but no statistically significant connection to a clinical parameter was found. Being such notable targets of MSI is consistent with the notion that MSI in these loci is a hallmark of biochemical events that advance thyroid cancer.

There is good reason to suspect that MSI in D2S123 is fundamentally connected to follicular cancer, given the level of statistical significance associated with the high (100%) occurrence of MSI in D2S123. From our analysis of the four microsatellites on chromosome 2, it appears noteworthy that the spectrum of MSI is quite sharp, because there was no (0%) occurrence of MSI at the nearby BAT-26, D2S399, or D2S115 loci in the seven follicular cancer cases we analyzed. For the entire panel of 76 thyroid tumors, the MSI frequencies at BAT-26 (8.3%), D2S123 (32.4%), D2S399 (14.3%), and D2S115 (22.4%) compared with their chromosome 2 position of 49330585, 53180120, 174313142, and 206995452, respectively (UCSC Genome Browser October 7, 2000 Freeze),6 indicates that in sporadic thyroid cancer MSI varies nonuniformly in this portion of chromosome 2. Given the need to better define genetic differences in various forms of thyroid cancer, additional work on the genetic defect(s) that focus MSI at D2S123 in follicular cancer is, we feel, warranted.

The frequency of MSI in TSHR was approximately half that measured in THRA1; however, MSI in TSHR had stronger ties to clinical data than THRA1, being connected to patient age and nodal involvement. The TSHR gene appears to be important in thyroid cancer, because point mutations in TSHR are often found in thyroid carcinomas (34). The TSHR microsatellite we analyzed resides in a noncoding segment of the gene; consequently, it is unclear exactly how MSI in this locus would manifest itself clinically. We postulate that MSI in the TSHR microsatellite could affect gene expression in a number of ways, for example, by affecting changes in the degree of supercoiling and therein altering transcriptional regulation of the gene (35, 36) or by precipitating alternatively spliced variants of the gene (37). It is also reasonable to expect that MSI in a microsatellite such as TSHR, residing in an intron, occurs in concert with MSI in distant microsatellites that reside in exons. There are a host of tumor suppressor genes with coding microsatellites, where MSI produces frameshift mutations that knock out gene expression (33, 38) Much additional work is clearly needed to delineate the effects and spectrum of MSI in cancer to determine how MSI in noncoding microsatellites manifests a clinical phenotype.

In our study THRA1, TSHR, and D2S123 distinguished themselves as being of special significance for an assessment of MSI in thyroid cancer. In the case of THRA1, the highest frequency of MSI was observed, and MSI in THRA1 was also highly indicative of MSI being observed in other microsatellites. In the case of TSHR and D2S123, MSI proved to be correlated with important clinical parameters. For these reasons any assessment of MSI in thyroid cancer should ideally include THRA1, TSHR, and D2S123. From our data MSI appears to be an integral part of thyroid carcinogenesis as evidenced by the high frequency on MSI and the strong correlation to clinical and prognostic data we observe. As the link between MSI and thyroid cancer becomes clearer the genetics of thyroid cancer will also be brought into better focus. This, in turn, will aid in pinpointing genetic targets on which to center potent anticancer therapies.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

        
1

Supported by the California Cancer Research Program, Breast Cancer Research Program Department of Energy, The Fukushima Society for the Promotion of Medicine, Danish Research Agency, and Danish-American Fulbright Commission.

                
3

The abbreviations used are: PTC, papillary thyroid carcinoma; MSI, microsatellite instability.

        
4

Internet address: http://www.ncbi.nlm.nih.gov/.

        
5

Internet address: http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi.

        
6

Internet address: http://genome.cse.ucsc.edu/index.html.

Fig. 1.

Autoradiographs of MSI in thyroid samples. PCR product of the indicated loci analyzed by denaturing PAGE (see text for additional details). Arrows, different sized PCR amplification products in tumor samples (T) compared with patient matched normal tissue (N).

Fig. 1.

Autoradiographs of MSI in thyroid samples. PCR product of the indicated loci analyzed by denaturing PAGE (see text for additional details). Arrows, different sized PCR amplification products in tumor samples (T) compared with patient matched normal tissue (N).

Close modal
Fig. 2.

The relationship between MSI phenotype and survival rate. MSI-positive group showed the better prognosis than the MSI negative group (P = 0.007). Survival curve was made with Kaplan-Meier method and tested by log-rank test.

Fig. 2.

The relationship between MSI phenotype and survival rate. MSI-positive group showed the better prognosis than the MSI negative group (P = 0.007). Survival curve was made with Kaplan-Meier method and tested by log-rank test.

Close modal
Table 1

Patient and clinicopathological data

Total 76 
Gender  
 Male 11 
 Female 65 
Age  
 Average (range) 55.1 (20–87) years 
Histological type  
 Papillary carcinoma 65 (85.6%) 
 Follicular carcinoma 7 (9.2%) 
 Anaplastic carcinoma 2 (2.6%) 
 Medullary carcinoma 2 (2.6%) 
Stage  
 Stage I 30 
 Stage II 
 Stage III 27 
 Stage IV 
 Unknown 
Prognosis  
 Survivor 65 
 Dead 10 
 Unknown 
Average of follow-up period (range) 166.8 (7–328) weeks 
Total 76 
Gender  
 Male 11 
 Female 65 
Age  
 Average (range) 55.1 (20–87) years 
Histological type  
 Papillary carcinoma 65 (85.6%) 
 Follicular carcinoma 7 (9.2%) 
 Anaplastic carcinoma 2 (2.6%) 
 Medullary carcinoma 2 (2.6%) 
Stage  
 Stage I 30 
 Stage II 
 Stage III 27 
 Stage IV 
 Unknown 
Prognosis  
 Survivor 65 
 Dead 10 
 Unknown 
Average of follow-up period (range) 166.8 (7–328) weeks 
Table 2

PCR primer setsa

LocusForward primerReverse primer
D2S115 CAGCCATATTGACTTGAACGTAAG GGTACAGCCCATGTGTGAGA 
D2S123 AATGGACAAAAACAGGATGC CCCTTTCTGACTTGGATACC 
D2S399 TTCACATGGCAGACCTGATTAT GCTCAATGGGAAGTTATTGAATG 
BAT-26 CAGTCAGAGCCCTTAACCTTTT GCTCCTTTCTAAGCCTTCTTCAC 
RET AGGGCCTTGGTAATGTAGACCT CATCCTGGCTCAGAATAAACCT 
D11S912 TACTGCTTTGGGTATGCATATG GCTTTTTGTCTAGCCATGATTG 
TSHR CTTGCCTCCTCCTCAACAAC CGTGAGAGAACTTTCCTGGC 
THRA1 CTTAAGCAGTGGGGAACCTG ATAGCATTGCCTTCCCATGT 
p53 ACTGCCACTCCTTGCCCCATTC AGGGATACTATTCAGCCCGAGGTG 
LocusForward primerReverse primer
D2S115 CAGCCATATTGACTTGAACGTAAG GGTACAGCCCATGTGTGAGA 
D2S123 AATGGACAAAAACAGGATGC CCCTTTCTGACTTGGATACC 
D2S399 TTCACATGGCAGACCTGATTAT GCTCAATGGGAAGTTATTGAATG 
BAT-26 CAGTCAGAGCCCTTAACCTTTT GCTCCTTTCTAAGCCTTCTTCAC 
RET AGGGCCTTGGTAATGTAGACCT CATCCTGGCTCAGAATAAACCT 
D11S912 TACTGCTTTGGGTATGCATATG GCTTTTTGTCTAGCCATGATTG 
TSHR CTTGCCTCCTCCTCAACAAC CGTGAGAGAACTTTCCTGGC 
THRA1 CTTAAGCAGTGGGGAACCTG ATAGCATTGCCTTCCCATGT 
p53 ACTGCCACTCCTTGCCCCATTC AGGGATACTATTCAGCCCGAGGTG 
a

Each primers set generated originally based on the sequence information from GenBank except D11S912 and p53, Sequence direction: 5′ to 3′ (left to right).

Table 3

Frequency of MSI with significance of correlation to THRA1 locus

LocusPositive/testedaFrequencyComparison to THRA1 locus
D2S115 17/76 22.4% NSb (P = 0.08) 
D2S123 24/74 32.4% NS (P = 0.72) 
D2S399 10/70 14.3% P = 0.004 
BAT-26 5/60 8.3% P = 0.0001 
RET 12/64 18.8% P = 0.03 
D11S912 21/76 27.6% NS (P = 0.32) 
TSHR 14/76 18.4% P = 0.021 
THRA1 27/74 36.5% — 
p53 13/76 17.1% P = 0.012 
LocusPositive/testedaFrequencyComparison to THRA1 locus
D2S115 17/76 22.4% NSb (P = 0.08) 
D2S123 24/74 32.4% NS (P = 0.72) 
D2S399 10/70 14.3% P = 0.004 
BAT-26 5/60 8.3% P = 0.0001 
RET 12/64 18.8% P = 0.03 
D11S912 21/76 27.6% NS (P = 0.32) 
TSHR 14/76 18.4% P = 0.021 
THRA1 27/74 36.5% — 
p53 13/76 17.1% P = 0.012 
a

MSI = positive case/tested informative case.

b

NS, not significant.

We thank Yayoi Mori, Masao Kato, Teiko Suzuki and Miyuki Mashiko for sample preparation. We also thank Shonna Hyde for preparing the manuscript.

1
Hundahl S. A., Cady B., Cunningham M. P., Mazzaferri E., McKee R. F., Rosai J., Shah J. P., Fremgen A. M., Stewart A. K., Holzer S. Initial results from a prospective cohort study of 5583 cases of thyroid carcinoma treated in the united states during 1996. U.S. and German Thyroid Cancer Study Group. An American College of Surgeons Commission on Cancer Patient Care Evaluation study.
Cancer (Phila.)
,
89
:
202
-217,  
2000
.
2
Fagin J. A. Molecular genetics of human thyroid neoplasms.
Annu. Rev. Med.
,
45
:
45
-52,  
1994
.
3
Halachmi N., Halachmi S., Evron E., Cairns P., Okami K., Saji M., Westra W. H., Zeiger M. A., Jen J., Sidransky D. Somatic mutations of the PTEN tumor suppressor gene in sporadic follicular thyroid tumors.
Genes Chromosomes Cancer
,
23
:
239
-243,  
1998
.
4
Pacini F., Elisei R., Romei C., Pinchera A. RET proto-oncogene mutations in thyroid carcinomas: clinical relevance.
J. Endocrinol. Investig.
,
23
:
328
-338,  
2000
.
5
Komminoth P. The RET proto-oncogene in medullary and papillary thyroid carcinoma. Molecular features, pathophysiology and clinical implications.
Virchows Arch.
,
431
:
1
-9,  
1997
.
6
Eng C. RET proto-oncogene in the development of human cancer.
J. Clin. Oncol.
,
17
:
380
-393,  
1999
.
7
Ain K. B. Anaplastic thyroid carcinoma: behavior, biology, and therapeutic approaches.
Thyroid
,
8
:
715
-726,  
1998
.
8
Vecchio G., Santoro M. Oncogenes and thyroid cancer.
Clin. Chem. Lab. Med.
,
38
:
113
-116,  
2000
.
9
Cahill D. P., Kinzler K. W., Vogelstein B., Lengauer C. Genetic instability and darwinian selection in tumours.
Trends Cell Biol.
,
9
:
M57
-M60,  
1999
.
10
Wirtz H. C., Muller W., Noguchi T., Scheven M., Ruschoff J., Hommel G., Gabbert H. E. Prognostic value and clinicopathological profile of microsatellite instability in gastric cancer.
Clin. Cancer Res.
,
4
:
1749
-1754,  
1998
.
11
Wright C. M., Dent O. F., Barker M., Newland R. C., Chapuis P. H., Bokey E. L., Young J. P., Leggett B. A., Jass J. R., MacDonald G. A. Prognostic significance of extensive microsatellite instability in sporadic clinicopathological stage C colorectal cancer.
Br. J. Surg.
,
87
:
1197
-1202,  
2000
.
12
Gryfe R., Kim H., Hsieh E. T., Aronson M. D., Holowaty E. J., Bull S. B., Redston M., Gallinger S. Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer.
N. Engl. J. Med.
,
342
:
69
-77,  
2000
.
13
Cravo M., Lage P., Albuquerque C., Chaves P., Claro I., Gomes T., Gaspar C., Fidalgo P., Soares J., Nobre-Leitao C. BAT-26 identifies sporadic colorectal cancers with mutator phenotype: a correlative study with clinico-pathological features and mutations in mismatch repair genes.
J. Pathol.
,
188
:
252
-257,  
1999
.
14
Kinzler K. W., Vogelstein B. Lessons from hereditary colorectal cancer.
Cell
,
87
:
159
-170,  
1996
.
15
Lynch H. T., Lynch J. F. Hereditary nonpolyposis colorectal cancer.
Semin. Surg. Oncol.
,
18
:
305
-313,  
2000
.
16
Vermiglio F., Schlumberger M., Lazar V., Lefrere I., Bressac B. Absence of microsatellite instability in thyroid carcinomas.
Eur. J. Cancer
,
1
:
128
1995
.
17
Soares P., dos Santos N. R., Seruca R., Lothe R. A., Sobrinho-Simoes M. Benign and malignant thyroid lesions show instability at microsatellite loci.
Eur. J. Cancer
,
33
:
293
-296,  
1997
.
18
Lazzereschi D., Palmirotta R., Ranieri A., Ottini L., Veri M. C., Cama A., Cetta F., Nardi F., Colletta G., Mariani-Costantini R. Microsatellite instability in thyroid tumours and tumour-like lesions.
Br. J. Cancer
,
79
:
340
-345,  
1999
.
19
Dobosz T., Lukienczuk T., Sasiadek M., Kuczynska A., Jankowska E., Blin N. Microsatellite instability in thyroid papillary carcinoma and multinodular hyperplasia.
Oncology
,
58
:
305
-310,  
2000
.
20
Boland C. R., Thibodeau S. N., Hamilton S. R., Sidransky D., Eshleman J. R., Burt R. W., Meltzer S. J., Rodriguez-Bigas M. A., Fodde R., Ranzani G. N., Srivastava S. A National Cancer Institute Workshop on microsatellite instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer.
Cancer Res.
,
58
:
5248
-5257,  
1998
.
21
Hoang J. M., Cottu P. H., Thuille B., Salmon R. J., Thomas G., Hamelin R. BAT-26, an indicator of the replication error phenotype in colorectal cancers and cell lines.
Cancer Res.
,
57
:
300
-303,  
1997
.
22
Kitamura Y., Shimizu K., Tanaka S., Ito K., Emi M. Allelotyping of anaplastic thyroid carcinoma: frequent allelic losses on 1q, 9p, 11, 17, 19p, and 22q.
Genes Chromosomes Cancer
,
27
:
244
-251,  
2000
.
23
Fogelfeld L., Bauer T. K., Schneider A. B., Swartz J. E., Zitman R. p53 gene mutations in radiation-induced thyroid cancer.
J. Clin. Endocrinol. Metab.
,
81
:
3039
-3044,  
1996
.
24
Wyllie F. S., Haughton M. F., Rowson J. M., Wynford-Thomas D. Human thyroid cancer cells as a source of iso-genic, iso-phenotypic cell lines with or without functional p53.
Br. J. Cancer
,
79
:
1111
-1120,  
1999
.
25
Laudet V., Begue A., Henry-Duthoit C., Joubel A., Martin P., Stehelin D., Saule S. Genomic organization of the human thyroid hormone receptor α (c-erbA-1) gene.
Nucleic Acids Res.
,
19
:
1105
-1112,  
1991
.
26
Henderson B. E., Feigelson H. S. Hormonal carcinogenesis.
Carcinogenesis
,
21
:
427
-433,  
2000
.
27
Yoshida A., Asaga T., Masuzawa C., Kawahara S. Alteration of tumorigenicity in undifferentiated thyroid carcinoma cells by introduction of normal chromosome 11.
J. Surg. Oncol.
,
55
:
170
-174,  
1994
.
28
Huang Y., Bove B., Wu Y., Russo I. H., Yang X., Zekri A., Russo J. Microsatellite instability during the immortalization and transformation of human breast epithelial cells in vitro.
Mol. Carcinog.
,
24
:
118
-127,  
1999
.
29
Hermanek P., Sobin L. H., Wittekind C. How to improve the present TNM staging system.
Cancer (Phila.)
,
86
:
2189
-2191,  
1999
.
30
Onda M., Abe R., Tsuchiya A., Fukushima T., Ando Y., Yoshida T. K-ras, and p53 gene mutations in colorectal cancer in 57 Japanese patients.
Fukushima J. Med. Sci.
,
43
:
29
-39,  
1997
.
31
Zhu Y. M., Das-Gupta E. P., Russell N. H. Microsatellite instability and p53 mutations are associated with abnormal expression of the MSH2 gene in adult acute leukemia.
Blood
,
94
:
733
-740,  
1999
.
32
Damm K., Thompson C. C., Evans R. M. Protein encoded by v-erbA functions as a thyroid-hormone receptor antagonist.
Nature (Lond.)
,
339
:
593
-597,  
1989
.
33
Rampino N., Yamamoto H., Ionov Y., Li Y., Sawai H., Reed J. C., Perucho M. Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype.
Science (Wash. DC)
,
275
:
967
-969,  
1997
.
34
Russo D., Arturi F., Schlumberger M., Caillou B., Monier R., Filetti S., Suarez H. G. Activating mutations of the TSH receptor in differentiated thyroid carcinomas.
Oncogene
,
11
:
1907
-1911,  
1995
.
35
Yoshimura S. H., Ohniwa R. L., Sato M. H., Matsunaga F., Kobayashi G., Uga H., Wada C., Takeyasu K. DNA phase transition promoted by replication initiator.
Biochemistry
,
39
:
9139
-9145,  
2000
.
36
Kahl B. F., Li H., Paule M. R. DNA melting and promoter clearance by eukaryotic RNA polymerase I.
J. Mol. Biol.
,
299
:
75
-89,  
2000
.
37
Libri D., Goux-Pelletan M., Brody E., Fiszman M. Y. Exon as well as intron sequences are cis-regulating elements for the mutually exclusive alternative splicing of the β tropomyosin gene.
Mol. Cell. Biol.
,
10
:
5036
-5046,  
1990
.
38
Markowitz S., Wang J., Myeroff L., Parsons R., Sun L., Lutterbaugh J., Fan R. S., Zborowska E., Kinzler K. W., Vogelstein B. Inactivation of the type II TGF-β receptor in colon cancer cells with microsatellite instability.
Science (Wash. DC)
,
268
:
1336
-1338,  
1995
.