SEARCH

SEARCH BY CITATION

Keywords:

  • atypical follicular thyroid adenoma;
  • follicular thyroid adenoma;
  • benign entity;
  • follicular thyroid carcinoma;
  • TERT promoter mutation;
  • telomerase

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

BACKGROUND

The telomerase reverse transcriptase (TERT) promoter mutations C228T and C250T have been found in many malignancies, including in thyroid carcinomas. However, it is unclear how early these mutations occur in thyroid tumorigenesis.

METHODS

The study included primary tumors from 58 patients initially diagnosed with follicular thyroid adenoma (FTA), a benign entity, 18 with atypical FTA (AFTA) having an uncertain malignant potential, and 52 with follicular thyroid carcinoma (FTC). Sanger sequencing was used to investigate the mutational status of the TERT promoter. Telomere length and TERT messenger RNA (mRNA) expression were determined using quantitative polymerase chain reaction (PCR). Telomerase activity was assessed using a Telomerase PCR enzyme-linked immunosorbent assay kit.

RESULTS

The C228T mutation was identified in 1 of 58 FTA (2%) and 3 of 18 AFTA (17%) samples. These 4 tumors all expressed TERT mRNA and telomerase activity, whereas the majority of C228T-negative adenomas lacked TERT expression (C228T versus wild-type, P = .008). The C228T mutation was associated with NRAS gene mutations (P = .016). The patient with C228T-mutated FTA later developed a scar recurrence and died of FTC, whereas none of the remaining 57 patients with FTA had recurrence. No recurrence occurred in 3 patients with AFTA who carried C228T during the follow-up period (36-285 months). Nine of the 52 FTCs (17%) exhibited the TERT mutation (8 of 9 C228T and 1 of 9 C250T), and the presence of the mutation was associated with shorter patient survival.

CONCLUSIONS

TERT promoter mutations may occur as an early genetic event in thyroid follicular tumors that have not developed malignant features on routine histopathological workup. Cancer 2014;120:2965–2979. © 2014 American Cancer Society.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

Human telomeres are nucleoprotein complexes that form protective caps on chromosome ends and are essential for genomic stability/integrity.[1-4] Normal somatic cells undergo progressive attrition of telomeric DNA with each round of division and enter a permanent growth arrest stage called senescence when their telomere shortens to a critical size that elicits DNA damage response.[1-4] Therefore, telomere erosion-mediated cellular senescence prevents unlimited cellular proliferation, which is believed to act as a potent tumor suppressor.[1] Overcoming of the senescence barrier by telomere stabilization is thus required to achieve infinite cell proliferation, and activation of telomerase, an RNA-dependent DNA polymerase responsible for elongating telomere, plays a key role in oncogenesis.[1-5] Telomerase activity is detectable in up to 90% of human malignancies but is absent in most normal human cells.[2, 3] Telomerase reverse transcriptase (TERT) is a catalytic component of the telomerase complex and a key determinant for controlling telomerase activity.[2] Telomerase activation and silencing generally results from active or repressive transcription of the TERT gene, respectively, and the enzymatic activity is thus correlated with TERT expression. Numerous clinical observations have shown that telomerase and TERT contribute not only to sustained cell proliferation, but also to aggressive diseases and poor outcomes in many types of human malignancies.[2, 6, 7]

Given the important role of telomerase or TERT in cancer formation and progression, great efforts have been made to dissect the mechanisms behind telomerase activation and TERT gene regulation. Recently, the TERT promoter mutations C228T and C250T were identified in various human malignancies including thyroid carcinomas and shown to enhance TERT gene transcription or telomerase activation.[8-24] These mutations have been suggested as oncogenic driver events,[9, 10] putatively creating a binding site for ETS transcription factors, which in turn may be upregulated by the MAPK pathway activated through BRAF and NRAS mutations. The TERT promoter mutation is very frequent in anaplastic thyroid carcinoma (ATC) and frequent in follicular thyroid carcinoma (FTC) and papillary thyroid carcinoma (PTC).[12, 14, 16, 18, 20, 21] It has also been associated with increased age at diagnosis, and poor survival in PTC and FTC.[16, 18, 20, 21] However, it is currently unclear how early this genetic event occurs in oncogenesis, or more broadly, it remains to be defined how early telomerase is activated in carcinogenesis, especially in an in vivo setting. Because the TERT promoter mutation is not present in normal tissues or cells, it might serve as an ideal genetic marker to monitor telomerase activation during the process of tumor development.

Follicular thyroid adenoma (FTA) is a benign entity, and atypical FTA (AFTA) is regarded as having an uncertain malignant potential.[25-27] Although the vast majority of FTA and most AFTA remain at a benign stage for a lifetime period, some patients recur with the malignant counterpart FTC.[25-32] Molecular and genetic studies demonstrate that certain important oncogenic alterations occurring in FTC are observed in FTA and/or AFTA, for instance, gain-of-function mutations in the RAS gene family and the PAX8/PPARγ fusion.[28-31] Given the proposed role of TERT promoter mutation in carcinogenesis and the presence of this mutation in follicular cell–derived thyroid carcinomas,[12, 14, 16, 18, 20, 21] we thus determined whether the mutation is present in primary tumors diagnosed as FTA or AFTA, and if so, whether this leads to telomerase activation.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

Patients and Tumor Specimens

The study included primary tumor specimens from 76 patients who were diagnosed and operated in the period from 1986 to 2004 at the Karolinska University Hospital Solna, Sweden. The primary tumors were diagnosed as follicular thyroid adenomas according to the criteria of the World Health Organization (WHO).[25-27] FTA is defined as “A benign encapsulated tumor of the thyroid showing evidence of follicular cell differentiation” and “By definition, capsular or vascular invasion is absent”.[25-27] A subset of 18 cases was classified as AFTA based on the presence of the following features: high cellularity, signs of increased proliferation ie, mitosis in routinely stained slides, and/or aberrant relation between the tumor and the capsule or vessels without fulfilling criteria for an FTC diagnosis. Altogether, 58 tumors were classified as FTA (22 male and 36 female) and 18 as AFTA (2 male and 16 female). The median age at diagnosis was 48 years (range, 23-88 years) for FTA cases and 49 years (range, 25-72 years) for AFTA. Patients were followed up for disease recurrence and survival for a median of 236 months (range, 67-316 months) in FTA patients and a median of 210 months (range, 32-298 months) in patients with AFTA. The detailed patient information is given in Tables 1 and 2.

Table 1. Mutations and Follow-Up for the 58 Patients With a Primary FTA
Case No.MutationAge at Diagnosis, ySex (M/F)Primary TumorFollow-UpFinal Diagnosis
TERT PromoterRAS GeneDisease RecurrencePatient OutcomeTime, mo
  1. AWOD, alive without disease; DWOD, dead without disease; DOD, dead of disease; wt, wild type; –, not determined.

FTA-1wt55FFTAnoDWOD172FTA
FTA-2wt40FFTAnoAWOD316FTA
FTA-3wt52MFTAnoDWOD87FTA
FTA-4wt32FFTAnoAWOD314FTA
FTA-5wt46FFTAnoAWOD313FTA
FTA-6wt40MFTAnoDWOD277FTA
FTA-7wt46MFTAnoAWOD309FTA
FTA-8wt50FFTAnoAWOD309FTA
FTA-9wt25MFTAnoDead281FTA
FTA-10wt61FFTAnoDWOD142FTA
FTA-11wt55FFTAnoDWOD254FTA
FTA-12wt50FFTAnoAWOD306FTA
FTA-13wt32MFTAnoAWOD305FTA
FTA-14wt64FFTAnoDWOD294FTA
FTA-15wt37FFTAnoAWOD304FTA
FTA-16wt62FFTAnoAWOD303FTA
FTA-17wt43MFTAnoAWOD303FTA
FTA-18wt54FFTA noAWOD303FTA
FTA-19wt49MFTAnoAWOD302FTA
FTA-20wt78FFTAnoDWOD198FTA
FTA-21C228TNRAS Q61R69FFTAyes, FTCDOD250FTC
FTA-22wt52MFTAnoAWOD301FTA
FTA-23wt50MFTAnoAWOD301FTA
FTA-24wt63MFTAnoDWOD255FTA
FTA-25wt57FFTAnoAWOD297FTA
FTA-26wt43MFTAnoAWOD296FTA
FTA-27wt52FFTAnoDWOD289FTA
FTA-28wt61FFTAnoDWOD193FTA
FTA-29wt47FFTAnoAWOD293FTA
FTA-30wt36MFTAnoDWOD288FTA
FTA-31wt77MFTAnoDWOD67FTA
FTA-32wt38MFTAnoAWOD289FTA
FTA-33wt27FFTAnoAWOD286FTA
FTA-34wt62FFTAnoDWOD95FTA
FTA-35wt-52MFTAnoAWOD285FTA
FTA-36wt-44FFTAnoAWOD284FTA
FTA-37wt-55FFTAnoAWOD281FTA
FTA-38wt-48FFTAnoAWOD280FTA
FTA-39wt-36MFTAnoAWOD274FTA
FTA-40wt-67FFTAnoDWOD166FTA
FTA-41wt-43MFTAnoAWOD274FTA
FTA-42wt-34FFTAnoAWOD273FTA
FTA-43wt33FFTAnoAWOD273FTA
FTA-44wt45MFTAnoAWOD272FTA
FTA-45wt44MFTAnoAWOD271FTA
FTA-46wt49FFTAnoAWOD256FTA
FTA-47wt47MFTAnoAWOD247FTA
FTA-48wt29FFTAnoAWOD225FTA
FTA-49wt33FFTAnoAWOD185FTA
FTA-50wt79FFTAnoAWOD127FTA
FTA-51wt88FFTAnoDWOD53FTA
FTA-52wt41FFTAnoAWOD115FTA
FTA-53wt36MFTAnoAWOD112FTA
FTA-54wt60MFTAnoAWOD111FTA
FTA-55wt51FFTAnoAWOD111FTA
FTA-56wt47FFTAnoAWOD109FTA
FTA-57wt61FFTAnoAWOD109FTA
FTA-58wt23FFTAnoAWOD108FTA
Table 2. Mutations and Follow-Up for the 18 Patients With a Primary Tumor Classified as AFTA
Case No.MutationAge at Diagnosis, ySex (M/F)Histopathology of Primary TumorPrimary Tumor ClassificationFollow-UpFinal Diagnosis
TERT PromoterRAS GeneSize, cmHigh CellularityMitoses ObservedAberrant Relation ToDisease RecurrencePatient OutcomeTime, mo
CapsuleVessel
  1. a

    Based on total gland size.

  2. Abbreviations: AWOD, alive without disease; DWOD, dead without disease; DOD, dead of disease; wt, wild type; –, not observed.

AFTA-1wtwt72F3.5cellularitymitosesAFTAnoDWOD166AFTA
AFTA-2wtwt52F1.5cellularitycapsuleAFTAnoAWOD298AFTA
AFTA-3C228TNRAS Q61K 45F2.5cellularitymitosesAFTAnoAWOD285AFTA
AFTA-4wtwt36F3cellularitymitosescapsuleAFTAnoAWOD284AFTA
AFTA-5wtwt70F6cellularitymitosescapsulevesselAFTAnoDWOD78AFTA
AFTA-6wtwt49F3cellularitymitosesAFTAnoAWOD277AFTA
AFTA-7wtwt49F2cellularitymitosescapsuleAFTAnoAWOD277AFTA
AFTA-8C228Twt60F4.5cellularitymitosesAFTAnoDWOD32AFTA
AFTA-9wtwt36M4.5cellularitycapsuleAFTAnoAWOD259AFTA
AFTA-10wtwt47F4cellularitymitosesAFTAnoAWOD259AFTA
AFTA-11wtwt43M<5.5acapsuleAFTAnoAWOD237AFTA
AFTA-12wtwt25F2cellularitycapsuleAFTAnoAWOD236AFTA
AFTA-13wtwt48F3cellularitymitosescapsuleAFTAnoAWOD224AFTA
AFTA-14wtwt63F<8 *cellularitymitosesvesselAFTAyes, FTCDOD139FTC
AFTA-15wtNRAS Q61R33F3.5capsuleAFTAnoAWOD195AFTA
AFTA-16wtwt37F5cellularitymitosesvesselAFTAnoAWOD184AFTA
AFTA-17C228TNRAS Q61R49F2cellularitymitosescapsuleAFTAnoDWOD181AFTA
AFTA-18wtNRAS Q61R54F3.5cellularitymitosescapsuleAFTAnoAWOD165AFTA

For comparison, surgical specimens from 20 patients with a nontumorous thyroid lesion were studied. Furthermore, primary tumors from 52 cases of FTC were studied (Table 3), 36 of which were previously reported.[18] The specimens were collected after surgical treatment within the Karolinska University Hospital Biobank and kept frozen at −70°C until use. All samples were collected with informed consent and approval by the local ethics committee.

Table 3. TERT Promoter Mutation and Clinical Details for the 52 Patients With FTCa
Case No.TERT Promoter MutationAge at Diagnosis, ySex (M/F)Primary Tumor DiagnosisFollow-Up
Clinical OutcomePatient SurvivalTime, mo
  1. a

    Results for 36 of the cases were reported in Liu et al.[18]

  2. Abbreviations: AWOD, alive without disease; DOD, dead of disease; DWOD, dead without disease; FTC, follicular thyroid cancer; F, female; M, male; MI, minimally invasive; WI, widely invasive.

FTC-1wt32FFTC-MIAWODAlive321
FTC-2wt47MFTC-WIDODDead31
FTC-3C228T31MFTC-MIAWODAlive282
FTC-4wt70FFTCDODDead29
FTC-5C228T57MFTC-MIDODDead160
FTC-6wt77FFTC-WIDWODDead26
FTC-7wt57MFTC-MIDODDead193
FTC-8wt49MFTC-MIAWODAlive246
FTC-9wt55FFTC-MIAWODAlive241
FTC-10wt49FFTC-MIAWODAlive241
FTC-11wt79FFTC-WIDODDead57
FTC-12wt74MFTC-WIDWODDead36
FTC-13wt43FFTC-WIAWODAlive120
FTC-14wt52FFTC-WIAWODAlive120
FTC-15wt48MFTC-WIDODDead42
FTC-16wt80MFTC-WI  29
FTC-17wt45FFTC-WIAWODAlive111
FTC-18wt65MFTC-WIAWODAlive111
FTC-19wt74FFTC-WIDWODDead37
FTC-20wt14FFTC-WIAWODAlive101
FTC-21C228T61MFTC-WIDODDead62
FTC-22wt30FFTC-WIAWODAlive66
FTC-23wt15FFTC-WIAWODAlive47
FTC-24wt86MFTC-WIDWODDead7
FTC-25wt51FFTC-WIAWODAlive34
FTC-26wt55FFTC-WIAWODAlive7
FTC-11wt57FFTC-MIAWODAlive238
FTC-12C250T81FFTC-WIDODDead71
FTC-13wt44MFTC-MIAWODAlive230
FTC-14C228T83FFTC-MIDWODDead126
FTC-15wt24FFTC-MIAWODAlive221
FTC-16C228T66FFTC-WIDODDead109
FTC-17wt75FFTC-MIDWODDead137
FTC-18wt76MFTC-MIDODDead40
FTC-19wt60FFTC-WIDODDead34
FTC-20wt50FFTC-MIAWODAlive207
FTC-21wt41FFTC-MIAWODAlive185
FTC-22C228T81MFTC-MIDWODDead112
FTC-23wt56FFTC-MIAWODAlive190
FTC-24wt31FFTC-MIAWODAlive174
FTC-25wt24FFTC-MIAWODAlive171
FTC-26wt70FFTC-MIDWODDead67
FTC-27wt50MFTC-MIDWODDead21
FTC-28wt76FFTC-MIAWODAlive180
FTC-29wt67FMTC-FTCAWODAlive158
FTC-30wt45FFTC-MIAWODAlive169
FTC-31C228T72MFTC-MIDODDead41
FTC-32wt51MFTC-MIDODDead73
FTC-33C228T64MFTC-WIDODDead107
FTC-34wt52FFTC-WIAWODAlive109
FTC-35wt17FFTC-MIAWODAlive92
FTC-36wt65MFTC-WIAWODAlive100

DNA Extraction and Sequencing

Genomic DNA was extracted from frozen tissue samples and used for sequencing of the TERT promoter to detect the mutations C228T and C250T located at positions −124 and −146 bp upstream of the ATG (start codon) (Fig. 1A). The target region was amplified by PCR using primers detailed in Table 4, followed by Sanger sequencing.[18] Mutations were verified by sequencing in both directions. The FTA and AFTA samples were also analyzed for HRAS, NRAS, and KRAS mutations at the hotspot codons 12 and 13 in exon 1 and codon 61 in exon 2 by Sanger sequencing using primers described in Table 4.

image

Figure 1. Identification of TERT promoter mutations and NRAS gene mutations in follicular thyroid adenoma (FTA) and atypical FTA (AFTA). (A) Top: Location of C228T and C250T (in red) in the TERT core promoter. TSS: Transcription start site. Below: Sequencing chromatographs of the TERT promoter locus in genomic tumor DNA from FTA-21 (left) and AFTA-3 (right) obtained by Sanger sequencing. C-to-T transitions at C228 in the TERT promoter locus is shown for both cases. (B) Sequencing chromatographs showing an A[RIGHTWARDS ARROW]G missense mutation Q61R at codon 61 of the NRAS gene in FTA-21 (left) AFTA-3 (right).

Download figure to PowerPoint

Table 4. Details of the Primers Used in the Present Study
Purpose/TargetDescriptionSequence
Sequencing of TERT  
TERT promoterSense5′-CACCCGTCCTGCCCCTTCACCTT-3′
TERT promoterAntisense5′-GGCTTCCCACGTGCGCAGCAGGA-3′
Telomere length  
TelomereTel 1b5′-CGGTTTGTTTGGGTTTGGGTTTGGGTTTGGGTTTGGGTT-3′
TelomereTel 2b5′-GGCTTGCCTTACCCTTACCCTTACCCTTACCCTTACCCT-3′
HBG3 (β-globin) 5′-TGTGCTGGCCCATCACTTTG-3′
HBG3 (β-globin) 5′-ACCAGCCACCACTTTCTGATAGG-3′
Sequencing of HRAS  
HRAS exon 1 codon 12/13Forward5′-ATGACGGAATATAAGCTGGT-3′
HRAS exon 1 codon 12/13Reverse5′-CTCTATAGTGGGGTCGTATT-3′
HRAS exon 2 codon 61Forward5′-AGGTGGTCATTGATGGGGAG-3′
HRAS exon 2 codon 61Reverse5′-AGGAAGCCCTCCCCGGTGCG-3′
Sequencing of KRAS  
KRAS exon 1 codon 12/13Forward5′-GGCCTGCTGAAAATGACTGAA-3′
KRAS exon 1 codon 12/13Reverse5′-GGTCCTGCACCAGTAATATGC-3′
KRAS exon 2 codon 61Forward5′-CAGGATTCCTACAGGAAGCAAGTAG-3′
KRAS exon 2 codon 61Reverse5′-CACAAAGAAAGCCCTCCCCA-3′
Sequencing of NRAS  
NRAS exon 1 codon 12/13Forward5′-ATGACTGAGTACAAACTGGT-3′
NRAS exon 1 codon 12/13Reverse5′-CTCTATGGTGGGATCATATT-3′
NRAS exon 2 codon 61Forward5′-TCTTACAGAAAACAAGTGGT-3′
NRAS exon 2 codon 61Reverse5′-GTAGAGGTTAATATCCGCAA-3′

Relative Telomere Length

Telomere length was assessed using a PCR-based method[33, 34] with primers for human telomere (T) and the internal control β-globin (S) (Table 4). Mean relative telomere lengths were determined by calculating T/S values using the formula T/S = 2−ΔCt, where ΔCt = mean Cttelomere − mean Ctβ-globin.[33, 34]

Quantitative Real-Time PCR (qRT-PCR) for TERT mRNA Expression

Total cellular RNA was extracted using Trizol kit (Life Technology) and used for cDNA synthesis. qRT-PCR was performed using an ABI 7900HT real-time PCR System (Applied Biosystems) and TaqMan Gene Expression Assays (Applied Biosystems) for TERT (Hs00 972656_m1) and 18S rRNA (Hs99999901_s1). Expression levels of TERT were calculated from threshold cycle values and normalized to 18S values.

Telomerase Activity

The TeloTAGGG Telomerase PCR ELISA kit (Roche Diagnostics GmbH, Mannheim, Germany) and 5 μg protein per sample were used for the telomerase activity assay. Reaction mixtures with lysis buffer and HEK-293 (human embryonic kidney) cell protein extracts were used as negative and positive controls, respectively. Telomerase activity was calculated from the absorbance (optical density) at 450 nm and 690 nm wavelength (OD450 and OD690, respectively) and expressed in arbitrary units.

Statistical Analyses

The Mann-Whitney U test was used to compare telomere lengths between patients with and those without the TERT promoter mutation. Differences in TERT expression between C228T-positive and -negative tumors and differences in the TERT promoter mutation frequency between tumors with different RAS mutation statuses or clinical parameters were determined using Fisher's exact test. Overall survival (OS) and disease-related survival (DRS) were illustrated by Kaplan-Meier plots, and significance was calculated by log-rank test. Cumulative hazard estimates were used to determine the impact of age and TERT promoter mutations on patients' overall and disease-related death risk. All the tests were 2-tailed and computed using SigmaStat3.1R software (Systat Software, Inc., Richmond, Calif). P values of < .05 were regarded as statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

TERT Promoter C228T Mutation as an Early Event in Follicular Thyroid Tumors

We screened for TERT promoter mutations in 76 primary tumor specimens from 58 patients originally diagnosed with FTA and 18 with AFTA (Table 1 and 2). The C228T mutation was identified in 1 of 58 FTAs (2%) and 3 of 18 AFTAs (17%) (Fig. 1A). However, no case exhibited the C250T mutation. As a comparison, we also analyzed 20 nontumorous thyroid specimens and they were all found to be mutation-negative. The results clearly demonstrate that the TERT promoter mutation C228T is an early genetic event that occurs in tumors without an histopathologically overt malignant phenotype (Fig. 2).

image

Figure 2. Photomicrographs showing histopathological characteristics of atypical thyroid adenoma cases AFTA-3, AFTA-8, and AFTA-17 with the TERT C228T mutation after staining by hematoxylin and eosin. Each case is shown in low magnification (×10, left) and high magnification (×100, right). All cases exhibit high cellularity and mitoses.

Download figure to PowerPoint

TERT Expression and Telomerase Activation in FTAs and AFTAs Carrying the TERT Promoter Mutation

The TERT promoter mutation is known to stimulate TERT transcription.[10] We thus determined whether the presence of C228T contributes to in vivo induction of TERT expression and telomerase activity in FTAs and AFTAs. TERT mRNA and telomerase activity were readily detectable in all four C228T-positive samples (Fig. 3A,B). Seven FTA and 11 AFTA tumors without TERT promoter mutation were also analyzed for TERT mRNA expression. Six of 7 FTAs did not have detectable TERT mRNA and only 1 of them expressed trace amounts of TERT transcripts (Fig. 3A). Among 11 examined AFTAs, 7 were totally negative whereas 2 were positive for TERT expression, and the remaining 2 exhibited negligible levels of TERT mRNA (Fig. 3B). Altogether, there was a highly significant difference in TERT expression status between C228T-positive and negative tumors (Fisher's exact test, P = .008) (Fig. 3C). The presence of TERT transcripts was well correlated with telomerase activation in the FTA and AFTA specimens examined (Fig. 3 and data not shown).

image

Figure 3. Induction of TERT mRNA expression coupled with telomerase activation in FTA and AFTA patients carrying the TERT promoter mutation. (A,B) Total RNA derived from the TERT promoter C228T mutation-positive and negative FTA (A) (n = 8) and AFTA (B) (n = 14) patients was analyzed for TERT mRNA expression using qRT-PCR and the relative mRNA level was expressed as arbitrary units. (C) Comparison of TERT expression status and TERT mutation status shows a highly significant difference in TERT expression positivity between C228T-positive and TERT promoter wild-type tumors (Fisher's exact test, P = .008). (D) Telomerase activity in FTA and AFTA tumors was determined using a Telomerase PCR-ELISA kit and expressed as absorbance in arbitrary units.

Download figure to PowerPoint

Correlation of the TERT Promoter Mutation with NRAS Gene Mutation

Activating RAS gene mutations are known to occur in both FTA and AFTA.[28] To observe whether there is a correlation between these 2 genetic events, we further screened all 18 AFTAs and the C228T-positive FTA for mutations at codons 12, 13 and 61 of the HRAS, KRAS and NRAS genes. All 19 tumors carried wild-type HRAS and KRAS genes. In contrast, 5 of them were shown to have an NRAS missense mutation at codon 61. One case showed a CAA-to-AAA transition at nucleotide 181, leading to a change from glutamine to lysine (Q61K), and 4 cases had a CAA-to-CGA transition at nucleotide 182, giving a change from glutamine to arginine (Q61R) (Fig. 1B). Three of the NRAS-mutated cases (1 FTA and 2 AFTA) were also positive for C228T (Tables 1 and 2). There was a significant difference in the presence of the C228T mutation between tumors with and without NRAS mutation (P = .016).

Telomere Length in FTAs/AFTAs With and Without the TERT Promoter Mutation

We determined the mean telomere lengths in all FTA and AFTA specimens and the results showed that their telomere lengths varied substantially. The C228T mutation-positive tumors tended to have shorter telomeres compared to the negative ones, but the difference was not statistically significant (3.97 ± 2.17 versus 5.32 ± 2.95, P = .373). In addition, there was no correlation between patients' age and telomere length in tumor tissues (data not shown).

Relation Between the TERT Promoter Mutation and Age/Sex

The age at diagnosis of the 76 FTA and AFTA patients ranged from 23 to 88 years with a median of 49 years. All 4 patients with the TERT promoter mutation were ≥ 45 years old at diagnosis, and their ages were 45, 49, 60, and 69, respectively. In contrast, none of 28 patients with age < 45 carried the mutation. However, a statistical analysis did not show a significant difference (< 45 versus ≥ 45, Fisher's exact test, P = .29) (Tables 1 and 2), likely due to the small number of patients with the TERT promoter mutation. In addition, all of these C228T-positive patients were female (4 of 52), in contrast to none in 24 males, however, there was no statistically significant difference (P = .31) (Tables 1 and 2).

Relation Between TERT Promoter Mutation and Recurrence With Malignant FTC

Follow-up of all 76 FTA and AFTA patients showed that a total of 2 patients had recurrence with FTC, whereas the other 74 patients remained disease-free (Tables 1 and 2). The C228T-positive FTA patient had a recurrence at the scar site 3 years after resection of the primary tumor. The primary tumor was a radically removed, 4 cm encapsulated tumor without high cellularity, mitosis or aberrant relation to the capsule or vessels. By contrast, the tumor in the recurrence showed high cellularity, nuclear size variation, and smaller areas with a suspected invasive growth pattern. The patient subsequently developed metastasis and died of the disease, thus confirming a final diagnosis of FTC. In contrast, none of 57 TERT mutation-negative FTA cases developed recurrence with FTC (Table 1). In the AFTA group, however, one TERT promoter mutation-negative AFTA developed recurrence with FTC (Table 2). This patient expressed high levels of TERT mRNA and telomerase activity despite the lack of the TERT promoter mutation (Fig. 3B,C; AFTA-14). The three C228T-carrying AFTA patients remained disease-free during follow-up.

The TERT Promoter Mutation in FTC

We then compared the findings in FTA and AFTA with TERT promoter sequencing analyses of 52 FTCs. Analyses of primary cancer tissues from 52 patients with FTC revealed 9 tumors (17%) that carried TERT promoter mutations (eight C228T and one C250T mutations, Table 3). The frequency was very similar to that in AFTA. The patients with the mutation were older at diagnosis than those without mutation and the difference was at borderline significance (66 ± 16 versus 54 ± 19, P = .052). There was no relation between the mutation and local invasiveness or metastasis. Univariate analyses showed that both the presence of TERT promoter mutations and age > 45 years were significantly associated with shorter OS and DRS (Fig. 4A,B and Fig. 4E,F; P < .05 and P < .01, respectively). On multivariate analysis, age > 45 remained significantly associated with shorter OS and DRS (Fig. 4D,H; P < .05), whereas patients carrying TERT promoter mutation stratified by age > 45 years tended to have shorter OS and DRS, but the difference was not significant (Fig. 4C,G; P = .192 and .099, respectively). However, the cumulative hazard estimate showed that the presence of TERT promoter mutations significantly increased patients' disease-related death risk (Fig. 5).

image

Figure 4. Overall survival (left) and disease-related survival (right) according to TERT promoter mutation statuses (mut and wt) and/or age (> 45 and ≤ 45 years) in patients with FTC. Kaplan-Meier plots analysis was performed on 51 patients with FTC with the follow-up information. (A and E) FTC patients according to the TERT promoter mutation. (B and F) FTC patients according to age at diagnosis > 45 or ≤ 45 years. (C and G) FTC patients according to TERT promoter mutation stratified by age cutoff 45 years. (D and H) FTC patients according to age > 45 or ≤ 45 years stratified by TERT promoter mutation status.

Download figure to PowerPoint

image

Figure 5. The cumulative hazard analyses of TERT mutation and age on follicular thyroid cancer (FTC) patients' death risk. Cumulative hazard estimates were performed on 51 patients with FTC to determine the impact of TERT promoter mutation and age on their overall and disease-related death risk. (A-C) For overall death risk analyses, and (D-F) for disease-related death risk. “Mut” and “wt” indicate patients with mutant and wild-type TERT promoter, respectively. Asterisk (*) indicates statistically significant. TERT promoter mutations were significantly associated with disease-related death risk, whereas age > 45 years increased overall death risk significantly.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

Despite the numerous studies showing widespread telomerase and TERT expression in human cancer cells, it is incompletely defined how early telomerase activation occurs in carcinogenesis, especially in an in vivo setting. This issue has been previously addressed by analyzing telomerase activity or TERT expression in human premalignant tumors,[27, 35] however, the obtained results are not totally reliable due to the following reasons. First, tumor tissue heterogeneity plus infiltration with telomerase-positive inflammatory cells may significantly interfere with telomerase/TERT assessments and interpretation of results. Second, specific TERT antibodies are not available.[36] The recently identified TERT promoter mutations should be an ideal genetic marker for telomerase activation, because the mutation stimulates TERT transcription and is only present in cancer cells.[9, 10] We thus determined the TERT promoter status in primary thyroid tumors classified as FTA or AFTA, a benign entity or entity of uncertain malignant potential, respectively. The TERT promoter mutation was observed in 1 of 58 FTAs (2%) and 3 of 18 AFTAs (17%), and was associated with increased TERT expression and telomerase activity providing direct evidence that telomerase activation can appear genetically during in vivo carcinogenesis already in tumors without histopathological evidence of a malignant phenotype. During preparation of this manuscript, Nault et al[19] reported that 25% of cirrhotic preneoplastic hepatic lesions harbored the TERT promoter mutation. Similarly, in bladder cancer, TERT promoter mutations were found to be an early genetic event present in noninvasive precursor lesions.[13]

The current findings would suggest that the AFTA entity is genetically similar to FTCs and different from FTAs without atypical features. The mutation frequency of 17% observed in AFTAs is comparable to the reports by us and others in differentiated thyroid cancers, ie, 14% to 36% in FTC and 8% to 27% in PTC, but lower than most observations of TERT mutations in undifferentiated ATC (13%-50%) and poorly differentiated thyroid cancer (PDTC; 21%-52%) (Table 5). Furthermore, the finding of a low TERT mutation frequency in FTA (2%) is overall in agreement with observations of 0% mutations in other reports (Table 5). Vinagre et al[16] examined 60 FTAs and did not find mutation-positive tumors. This is not surprising, given that this mutation appears as an infrequent genetic event in FTA. We also noticed that the patients included in the study were relatively younger than our cohort of patients (42 versus 48 years for median age), whereas age is one important factor associated with the TERT promoter mutation in both FTA and thyroid carcinomas.[18] In another study,[12] none of 85 benign thyroid tumors were positive for the TERT promoter mutation, however, the authors did not provide further information concerning tumor types, and clinical characters, and therefore it is difficult to evaluate or compare their data with the current results.

Table 5. Reports of TERT Promoter Mutations in Follicular Cell–Derived Thyroid Tumors
Tumor Type or SubgroupStudyNo. of Mutated/Analyzed Cases (%)
X Liu et alLanda et alVinagre et alT Liu et alX Liu et alMelo et alWang et al
(Ref [12])(Ref [14])(Ref [16])(Ref [18])(Ref [20])(Ref [21])(This study)
  1. a

    Also reported in the present study.

  2. Abbreviations: AFTA, atypical follicular thyroid adenoma; ATC, anaplastic thyroid carcinoma; FTA, follicular thyroid adenoma; FTC, follicular thyroid carcinoma; MI, minimally invasive; PDTC, poorly differentiated thyroid carcinoma; PTC, papillary thyroid carcinoma; WI, widely invasive.

ATC23/54 (43%)10/20 (50%)2/16 (13%)10/20 (50%)12/36 (33%)
PDTC3/8 (38 %)30/58 (52%)3/14 (21%)9/31 (29%)
PTC30/257 (12%)18/80 (23%)13/169 (8%)13/51 (26%)46/408 (11%)25/332 (8%)
FTC9/79 (12%)9/64 (14%)8/36 (22%)a8/22 (36%)12/70 (17%)9/52 (17%)
FTC, WI      4/24 (17%)
FTC, MI      5/26 (19%)
AFTA3/18 (17%)
FTA/benign0/85 (–)0/60 (–)0/44 (–)1/58 (2%)

The AFTA entity has been questioned due to a variety of criteria and there is currently no consensus definition in the WHO classification.[25-28, 30] Therefore, discrimination between AFTA and minimally-invasive FTC is not always performed in all institutions or based on uniform criteria. Here, all cases classified as AFTA exhibited high cellularity, mitosis, and/or aberrant relation between the tumor and the capsule or vessel (Table 2) but did not fulfil the criteria for FTC concerning capsular and/or vascular invasion (Fig. 2). All included AFTA cases were operated and diagnosed in the same institution, and the classification as AFTA was not based on uncertainty due to inadequate tumor sampling.

Many human malignancies follow a multistep model in oncogenesis, starting from benign, precursor lesions, and progressing to a full-blown carcinoma after varying lengths of time (several months to years).[27, 35] However, there is currently no consensus view of whether there exists an FTA/AFTA to FTC transition. It is commonly regarded that those patients who are initially diagnosed as FTA or AFTA but recur with FTC are due to misdiagnosis at the initial histopathological examination. On the other hand, there is also evidence that does support an evolution from FTA/AFTA to FTC.[29, 31] Likely, these 2 scenarios truly happen on different occasions. In this study, careful pathological reevaluation of all 4 TERT promoter-mutant patients was performed and the initial diagnosis was verified for the primary tumors (Fig. 2). The TERT promoter-mutated FTA initially presented without features of AFTA or FTC, but developed scar recurrence with histopathological features of AFTA and suggestive of FTC followed by recurrence with metastatic FTC. Either this case is an initially misclassified FTC with small areas of malignant features that were missed in the histopathological assessment, or it is a case of FTA with progression to FTC reflected in the alteration of histopathological and clinical phenotypes. Importantly, in both scenarios the FTC diagnosis was not detected in the primary tumor. The implication of this single FTA carrying a C228T mutation should be interpreted with caution. Identification of TERT promoter mutation may become a useful tool in follicular thyroid tumor diagnostic workup only if verified in additional cases and independent case series.

The 3 AFTAs and one FTA carrying the C228T mutation all expressed TERT mRNA and telomerase activity whereas the majority of the mutation-negative tumors lacked detectable TERT expression, suggesting that this genetic event does exert an in vivo stimulatory effect on TERT transcription and telomerase activation in early stages of oncogenesis. The finding is consistent with published observations showing the presence of TERT mRNA and telomerase activity in a fraction of FTA or AFTA patients.[28] However, one AFTA without TERT promoter mutation expressed the highest levels of TERT mRNA, which clearly indicates that other factors are operative in transactivating the TERT gene. It would be interesting to evaluate alterations such as TERT gene copy numbers, promoter methylation and regulatory effects as possible underlying mechanisms in this type of cases. Based on a lower frequency of their expression in FTA/AFTA, TERT and telomerase analyses may be applied in the differential diagnosis of thyroid cancer in fine-needle aspiration (FNA) biopsies. Some observations seem to support these two markers as useful adjuncts to FNA diagnosis of possible thyroid cancer, whereas others' results did not.[27, 31] Controversies likely result from tumor heterogeneity, infiltration of inflammatory cells, or lack of reliable TERT antibodies, as described above. Our findings demonstrated that the presence of the TERT promoter mutation was responsible for TERT/telomerase expression in most FTAs and AFTAs, and therefore simultaneous examination of both TERT promoter mutation and mRNA expression could be more informative and reliable.

In our study, the TERT promoter and NRAS mutations coexisted in 3 cases, 2 of which remained stable for the whole follow-up period (181 and 285 months, respectively). Clearly, both telomerase activation and NRAS mutation do not necessarily lead to metastatic FTC; however, a possible association to aggressive disease should be evaluated in additional cases. In this context it may be noted that Landa et al found an association between TERT promoter mutation and BRAF or RAS mutations in PDTC and ATC.[14]

In conclusion, we report the TERT promoter mutation in thyroid lesions classified as FTA, a benign entity, and AFTA having an uncertain malignant potential, which provides direct evidence that this genetic event can occur and subsequently activate telomerase before tumors acquire overt malignant phenotypes. In the FTA group, one patient with the mutation was found to develop FTC later; however, all 3 C228T-mutation–positive AFTA patients remained free from the original disease or FTC during the follow-up period. These results suggest that presence of the TERT promoter mutation, even coupled with RAS mutation, does not necessarily result in a malignant endpoint. Because the number of patients included in the present observation is limited, further studies are warranted to rigorously assess whether the mutation is a risk factor for FTC in larger cohorts of patients with thyroid adenomas.

FUNDING SUPPORT

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

Supported by The Swedish Cancer Society, the Swedish Research Council, the Cancer Society in Stockholm, the Stockholm County Council and Karolinska Institutet.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES