A gradually increased telomerase activity, assessed using the microdissection telomerase repeated amplification protocol (TRAP) and the measurement of mRNA hTERT expression, has been detected in the normal esophageal epithelium, dysplastic tissue carcinoma in situ (CIS) and esophageal squamous cell carcinoma (SCC) in two studies with clinical samples (15,16). There was a statistically significant difference between normal tissue and pre-neoplastic lesions (P<0.01), whereas no marked difference was found between pre-neoplasia and SCC (P>0.05) (15). Moreover, two studies investigated telomerase activity in iodine-non-reactive esophageal tissues. Those lesions, which remain unstained with Lugol's iodine staining, were related to inflammation, dysplasia and cancer development (17,18). By comparing telomerase activity between Lugol-stained and unstained epithelia using TRAP assay, it was concluded that the Lugol-unstained lesions presented a higher mean telomerase activity compared to the stained ones from the same patient. The unstained lesions included esophagitis, mild and severe dysplasia, and intramucosal and advanced SCC. Additionally, in the same study, the mRNA expression of hTERT was parallel to the increase of the atypia and malignant transformation (18).

Barret's esophagus (BE) is a pre-cancerous condition associated with esophageal adenocarcinoma (EAC). In a previous study, the methylation status of the promoters of the genes, hTERT, adenomatous polyposis coli (APC), TIMP metallopeptidase inhibitor 3 (TIMP3), cyclin-dependent kinase inhibitor 2A and secreted frizzled related protein 1, was compared between BE samples with EAC and those without EAC (19). The methylation rates for the first category were 92% for hTERT, 91% for TIMP3 and 100% for APC, while the other promoters reported no methylation. In BE without EAC, the same promoters were methylated in 17, 23 and 36% of the samples, respectively (P<0.0001). Therefore, the promoter methylation of hTERT appears to be involved in the carcinogenetic process of adenocarcinoma with pre-existing BE (19).

The mRNA expression of hTERT was assessed using polymerase chain reaction (PCR) in 60 cases of chronic gastritis, 15 of which presented intestinal metaplasia (IM) (20). The results revealed that hTERT mRNA expression was present in 23% of the cases, while the frequency for IM and non-IM samples was 47 and 16%, respectively (P=0.03) (20). Gastric carcinoma (GC) presented a hTERT positivity rate of 89%, with rates for well-differentiated and poorly differentiated tumors presenting 86 and 91%, respectively. hTERT expression was absent in the normal gastric mucosa (20). Wang et al (21) also found no hTERT expression in normal tissue samples, while the positivity rates for GC and pre-cancerous lesions were 87.5 and 47.4%, respectively (P<0.05). Finally, higher methylation rates of the hTERT promoter were observed in cancerous samples compared to the other groups (P<0.05), and there was no association between hTERT mRNA levels and Helicobacter pylori infection (20-22). Of note, gastric ulcer specimens presented an hTERT protein expression in 39% of the cases, while other studies reported no enzyme activity. In this case, hTERT upregulation could either contribute to the tissue's healing process or to its malignant transformation (22).

Gastric carcinogenesis and hTERT dysregulation are associated with the proto-oncogene MYC and the dysregulation of the TP53 tumor suppressor gene. The immunoreactivity of these genes was previously found to be higher in IM compared to superficial and atrophic gastritis samples (23). Additionally, the expression of the telomeric proteins telomere repeat factor (TRF)1, TRF2 and TERF1-interacting nuclear factor 2, which regulate TL, were found to vary among normal mucosa, pre-neoplastic tissue and GC specimens. Pre-cancerous lesions, GC, and GC with lymph node metastasis presented significantly higher levels of these proteins compared to normal tissue (P<0.01), while they were significantly increased in GC samples compared to pre-cancerous lesions (P<0.01). Finally, the mean TL was inversely related to the expression levels of the studied proteins. It was significantly shorter in the GC and GC samples with lymph node metastasis compared to the pre-cancerous lesions and normal tissue (P<0.01) (24).

It has been suggested that telomerase activation occurs during the progression from low- to high-grade dysplasia in adenomas, and increases progressively with the degree of dysplasia and invasion during colorectal carcinogenesis (25). It has been found that hTERT mRNA expression is a feature of the late-stage development of colorectal cancer (25). In a previous study, the expression of hTERT in normal colon mucosa from patients with advanced colorectal adenoma was evaluated and compared to that of the controls. The results did not reveal any difference between the two groups of patients (26). In another study, the level of hTERT mRNA expression in colorectal adenocarcinomas was significantly higher than that in corresponding non-tumorous mucosa tissues (P=0.009), and the expression level in the adenocarcinomas was slightly higher than that of adenomas, although the difference was not statistically significant (27). Of note, a higher level of hTERT expression was often noted in the adenocarcinomas arising from the left colon and rectum compared to those from the right colon (P=0.029) (27).

Only a limited number of studies have evaluated the association between colorectal cancer precursor lesions and TL. A previous study suggested that individuals with a short leukocyte TL had an increased risk of developing advanced adenomas (28). Roger et al (29) reported, in an experimental setting, that extensive tissue telomere erosion could lead to chromosomal instability and the initiation of colorectal cancer in polyps in patients with familial adenomatous polyposis. A recent study suggested that a short TL may be associated with an increased risk of developing colorectal polyps in both the adenoma-carcinoma and serrated pathways. That was the first study to report a statistically significant association between TL and serrated polyps, suggesting that telomeres may play an essential role along the entire serrated pathway (30).

In the study by Nault et al (31), the occurrence of hTERT promoter mutations during the malignant transformation of cirrhotic nodules into hepatocellular carcinoma (HCC) was evaluated. Their study included 58 patients with cirrhosis with HCC or pre-malignant lesions, including low-grade dysplastic nodules (LGDNs), high-grade dysplastic nodules (HGDNs), early HCC (eHCC), or small and progressed HCC (31). hTERT mutations were highly related to stepwise hepatocarcinogenesis, since they were identified in 6% of LGDNs, 19% of HGDNs, 61% of eHCCs, and 42% of small and progressed HCCs. There were 29 mutations which were detected in 96 nodules, including 25 cases mutated in the first hotspot at 2,124 base pairs (bp) before the ATG start (G>A substitution), and 4 cases mutated at the second known hotspot at 2146 bp before the ATG start (G>A substitution). These mutations were exclusive of each other. These hTERT promoter mutations were not found in the cirrhotic-matched tissues, indicating that they were somatic events (31).

In another study by the same research group, 401 liver samples from HCC, HCC lines, cirrhotic tissues, cirrhotic pre-neoplastic nodules and hepatocellular adenomas (HCAs) with or without malignant transformation were analyzed for hTERT promoter mutations (32). Mutations in hotspot regions were found in 179 cases (58%) of HCC and 15 cases (63%) of the HCC lines, indicating hTERT promoter mutations as the most frequent somatic genetic alterations in HCC. No mutation in the hTERT promoter was detected in the cirrhotic tissues. Among the 60 typical HCAs, no mutation in the hTERT promoter was detected. Of the 16 HCAs with malignant transformations, seven were positive for hTERT promoter mutations, indicating that hTERT promoter mutations were involved in the final step of the malignant transformation of HCA (32).

Matsuda et al (33) investigated TL between the normal pancreatic duct epithelium, pancreatic intraepithelial neoplasia (PanIN), and pancreatic cancer samples using in situ hybridization. PanIN types 1, 2 and 3, as well as cancer cells, exhibited weaker telomere signals in their nuclei and a decreased telomere centromere ratio (TCR) compared to the normal epithelium (P<0.05). Cancer cells also exhibited a lower TCR than the PanINs (P<0.05); however, it was unrelated to tumor grade. Atypical mitoses and anaphase bridges observed in PanIN and cancer cells were negatively associated with TCR (33). Based on the aforementioned findings, telomere shortening occurs early in pancreatic carcinogenesis and progresses as the malignant transformation develops (33).

Pre-neoplastic conditions of the biliary tract include hepatolithiasis, biliary epithelial hyperplasia and biliary epithelial dysplasia. Dysplastic changes are related to cases of chronic cholangitis, e.g., hepatolithiasis and primary sclerosing cholangitis and are considered to be a progenitor of intrahepatic cholangiocarcinoma (ICC). hTERT has been shown to be expressed in dysplastic cells and ICC samples, but not in hyperplastic cells and the normal bile duct epithelium (34). Furthermore, another study demonstrated that samples of acute or chronic inflammation of the gallbladder epithelium presented normal TLs (35). On the contrary, metaplastic lesions with pyloric or intestinal metaplasia presented shorter telomeres compared to normal cells (P<0.05) in 63% of the cases, whereas dysplastic cells and CIS in 91% of the cases. Notably, two CIS cases presented heterogeneity in TL, with both short and long telomeres in the nuclei. Lastly, cholangiocarcinoma, infiltrating adenocarcinoma of the gallbladder and extrahepatic bile ducts presented shorter telomeres in 98% of the cases (35).

Congenital malformations of the biliary tree also contribute to carcinogenesis through inflammation. Congenital biliary dilation (CBD) causes the dilation of the extrahepatic bile ducts and pancreaticobiliary maljunction (PBM), resulting in chronic inflammation of the biliary tract and gallbladder epithelium. Patients with CBD also have higher rates of cancer development in these areas. Using the Q-FISH assay, Aoki et al (36) calculated and compared the normalized TCR between CBD, cholecystolithiasis and normal tissue samples. All three categories exhibited significant differences with each other (P<0.001), and the TCRs for each tissue sample were 1.24, 1.96 and 1.77 for CBD, cholecystolithiasis and normal tissue, respectively, indicating that CBD cells presented shorter TLs (36). Moreover, in another study, non-cancerous samples from PBM gallbladders including chronic epithelial inflammation presented telomerase activity, with a score of 3.06, 32.95 and 17.93 total product generated (TPG) units in three different cases. On the other hand, PBM gallbladder carcinoma samples scored 46.57, 85.18, and 206.14 TPG units in three different cases. It is thus suggested that telomerase is a catalytic factor in PBM carcinogenesis (37).

The characterization of pre-neoplastic lesions of the lungs has been more comprehensively investigated in the case of squamous carcinoma, and to a lesser extent in adenocarcinoma and small-cell carcinoma. In the case of squamous carcinoma, it is generally considered that a stepwise accumulation of precursor phases occurs (38). The key pre-neoplastic lesions of the bronchial epithelium are atypical adenomatous hyperplasia (AAH), squamous dysplasia and CIS, as well as diffuse idiopathic pulmonary neuroendocrine cell hyperplasia. These are lesions of the bronchial epithelium and the precursors of lung adenocarcinoma, SCC and carcinoid tumors (39). Lantuejoul et al (40) performed an immunohistochemical and in situ hybridization study in pre-invasive and invasive bronchial lesions. They concluded that telomerase was increasingly expressed from the normal epithelium to squamous metaplasia, dysplasia and carcinoma in situ, and decreased in invasive carcinoma (P<0.0001), with a direct correlation between protein and mRNA levels of expression (P<0.0001). The expression of hTERT was also associated with resistance to apoptosis (40).

As regards TL, precancerous lung lesions present a shorter relative TL (RTL) than normal bronchial or alveolar tissue and invasive tumors. Thus, telomere shortening is considered an early event in lung cancer development, which precedes p53/Rb pathway inhibition and results in DNA damage responses (40,41). This could be achieved by activating shelterin complex components, such as the TERF1 and TERF2 proteins. These molecules stabilize the telomeres; however, their mRNA expression is increased in pre-neoplastic lesions, such as AAH. Lantuejoul et al (41) investigated the RTL of premalignant lung tissue using FISH and concluded that mild dysplastic lesions presented a lower RTL (RTL=1.2; normal cells, RTL=2), while this number increased in severe dysplasia, CIS and SCC (RTL=2). However, these differences were not statistically significant due to the low number of specimens studied. Similar proportions were observed from AAH to advanced ADC; AHH presented RTL=1.83, which increased in stage I-II ADC and stage III-IV ADC (2 and 1.88, respectively, P=0.047) (41). In addition, comparative genomic hybridization studies in samples from early stages of non-small cell lung cancer have shown that the genomic region that harbors the hTERT gene, 5p15.33, is frequently amplified compared with normal tissues (42).

An analysis of 56 pre-neoplastic breast tissues, including atypical ductal or lobular hyperplasia and lobular in situ carcinoma and comparison with healthy tissue and invasive carcinomas revealed that the pre-neoplastic lesions were more likely (60%) to have telomere shortening than normal breast tissue (35%; P=0.0116) (43). As regards invasive carcinomas, TLs were increased in 38.9% and markedly decreased in 38.9% of breast carcinomas (P=0.0087 for comparisons with pre-neoplastic lesions) (43). The telomere DNA content and the number of sites of allelic imbalance were assessed in a set of pre-invasive, invasive and healthy breast tissue samples. It was observed that the level of genomic instability did not differ between ductal carcinoma in situ and invasive carcinomas (44). In another study, telomerase activity was evaluated in 27 fibrocystic and dysplastic tissue samples, and 28 fibroadenomas and phylloid tumors, and was reported to be significantly increased in the dysplastic tissue and fibroadenoma groups compared to normal tissues (45).

A previous study with cervical samples from 100 patients revealed the overexpression of telomerase in 18.8% of the normal cervical samples, 32.0% of cervical intraepithelial neoplasia I (CIN I), 50.0% of CIN II, 60.0% of CIN III and 91.3% of invasive cervical cancer (46). Telomerase activity was significantly higher in patients with invasive cancer compared to those with CIN or a normal cervix (P<0.05), and its activity increased with the increasing CIN stage (46-48). Patients with benign lesions or CIN that exhibited TERC amplification relapsed or progressed to CIN II and CIN III more often than those without gene amplification (49-51). In addition, the levels of telomerase activity increased in parallel with the degree of CIN, with a significant increase during the transition to CIN3(52).

Importantly, there is an association between HPV E6 proteins and the activation of hTERT, resulting in increased risk of oncogenesis in the cervix (53). The human papillomavirus E6 protein binds to and activates the hTERT promoter of polymerase, promoting the precancerous transformation of the cervix. As previously demonstrated, based on testing 29 types of the virus, it was found that the oncogenic types specifically activate the hTERT promoter, while the non-oncogenic types do not. (53). The amplification of hTERT has been used in combination with HPV testing and cMYC amplification in order to optimize screening for malignancy in cytological samples (54).

The normal human endometrium expresses significant telomerase activity in a menstrual phase-dependent manner (55). In a previous study, a total of 32 normal endometrial tissues at various stages of menstruation or postmenopausal conditions were tested for telomerase expression (56). The expression of hTERT mRNA was characteristic in the normal endometrium and was dependent on the menstrual cycle phase. In the intrauterine proliferative phase, there was a rapid increase in the expression of hTERT, although not in the secretory phase. In additional to its expression in the normal endometrium, hTERT, hTR and TP1 were also found to be involved in the development of precancerous and cancerous lesions in the endometrium (56). Another study using a telomere-FISH assay to measure TLs compared chromosomal arm loss or gain in premalignant endometrial lesions with normal endometrium and reported TLs to be stable with the pathological transformation in endometrial hyperplasia and in endometrial carcinoma (57). A significantly increased number of telomere aggregates has been observed in atypical hyperplastic cells in mouse endometrial cancer models. It has been shown that alterations in the nuclear 3D telomere architecture are present in early proliferative lesions of mouse uterine tissues, indicative of endometrial cancer development (58).

In comparison with normal ovary/cystadenoma (32%), a previous study found a markedly higher frequency of moderate activity in low-malignant-potential tumors (67%) or invasive carcinomas (57%), suggesting a close association between the latter two categories (59). That study demonstrated a high prevalence of telomerase activity in low-malignant-potential tumors or invasive carcinomas, with the high telomerase activity associated exclusively with invasive ovarian carcinomas (59). In another study, telomerase activity assessed using TRAP assay was markedly increased in malignant compared to borderline tumors, benign tumors and normal ovaries (P<0.05) (60). The allelic discrimination analysis of primary and recurrent adult granulosa cell tumors has indicated that hTERT C228T promoter mutations are already present in some primary tumors; however, they may be late events that occur during adult granulosa cell tumor progression (61).

It has been suggested that hTERT promoter mutations represent the earlier onset of a clonal molecular process from which urothelial tumorigenesis may occur (62). hTERT promoter mutations were previously investigated among urothelial papilloma (UP) tumors and papillary urothelial neoplasms of low malignant potential (PUNLMPs). The results revealed that 46% of UPs and 43% of PUNLMPs carried a mutated hTERT promoter, and the activating hTERT promoter mutation C228T was detected in all mutant cases (63). These findings suggest that these low-malignancy entities share a transformation path similar to aggressive urothelial carcinomas (63). Another study investigated the significance of the hTERT promoter mutation pathway in the pathogenesis of inverted UP (IUP), an entity whose neoplastic nature is debated (64). The results demonstrated that 15% of inverted papillomas, 58% of urothelial carcinomas with inverted growth, 63% of conventional urothelial carcinomas and none of the cystic glandular specimens harbored a hTERT promoter mutation. It was suggested that a subset of inverted papilloma shares a developmental pathway similar to the carcinogenesis pathway of urothelial carcinoma. It has to be mentioned that a female predominance was noted in hTERT-mutated inverted papillomas, taking into account the strong male predilection of inverted papillomas in the general population. All hTERT promoter mutations were C228T, apart from two C250T mutations observed in two invasive urothelial carcinoma cases (64). However, another study conducted a comprehensive genetic analysis for multiple oncogenic genes in a sample size of 11 UPs and 11 IUPs. No IUP tumors had a mutated hTERT promoter. One UP tumor harbored a hTERT promoter mutation and was found in a patient with recurrent non-invasive papillary urothelial carcinomas (65). The results of that study are in contrast to those of previous studies on activating hTERT promoter mutations in IUP and UP, which may be attributed to different methodologies (63,64).

In another study, hTERT promoter mutations in cases of de novo PUNLMP were associated with a risk of recurrence (66). Recurrence with or without progression was encountered in 13 of 30 (43%) cases of PUNLMP, which were included. More specifically, 31% of the cases recurred as PUNLMP, 69% exhibited progression (54% progressed to non-invasive low-grade papillary urothelial carcinoma, 8% to non-invasive high grade papillary urothelial carcinoma and 8% developed stage progression to invasive high-grade urothelial carcinoma) (66). Among the recurrent tumors, 80% harbored a hTERT promoter mutation, including C250T and C228T, in contrast to 53% among the cases that did not recur (66).

Moreover, the presence of hTERT promoter mutations was previously assessed in a morphological spectrum of microdissected urothelia from urinary bladder specimens with and without keratinizing squamous metaplasia (KSM) and non-KSM (NKSM), including cases of neurogenic lower urinary tract dysfunction (NLUTD), and urothelial and squamous carcinomas (67). The results demonstrated that 94% of cancer foci, 68% of KSM and 70% of NKSM foci were positive for hTERT promoter mutations. The authors of that study suggested an association between conditions with chronic urinary bladder injury (such as NLUTD) and a higher risk of developing bladder cancer (67). In a recent study, hTERT promoter mutations were examined in whole-organ bladder samples, including cancerous tissue and samples of the tumor-associated normal urothelium, non-invasive urothelial lesions, carcinoma in situ and muscle-invasive bladder cancers (68). That study demonstrated that hTERT mutations were detected in tumor-associated normal urothelium and non-invasive urothelial lesions. Therefore, mutated hTERT promoter regions within non-invasive urothelial lesions are insufficient to establish cancerous growth, indicating the contribution of other gene mutations as a requirement for tumor development (68).

Pre-neoplastic lesions of the prostate gland include prostate intraepithelial neoplasia (PIN) and possibly atypical AAH, while benign conditions include benign prostate hyperplasia (BPH), which presents no risk for malignant transformation. Previously, when comparing BPH, PIN and prostate cancer telomeric fusion frequencies, the rates for each lesion were found to be similar and comparable: 65, 55 and 62%, respectively. The majority of normal prostatic epithelial tissue samples did not harbor telomeric fusions. As regards hTERT, all tissue samples presented detectable levels of its mRNA, with the rates being 69% in BPH, 60% in PIN and 94% in cancerous tissues. The normal adjacent epithelium also presented hTERT mRNA expression in 86% of the samples (69).

Variations are also evident in TL, as it appears that cancer telomeres are significantly shortened in comparison to normal tissue (P<0.05) (69). This characteristic is also shared by PIN, but not by BPH, as Southern blot analysis has revealed that normal epithelium and BPH have similar average TLs (6.6 and 6.4 kb, respectively) (70). Using Q-FISH assay, Cheng et al (71) compared TLs between normal epithelium, AAH, high-grade PIN and prostatic adenocarcinoma (PCA). Shortened telomeres were present in 20% of AAH, 68% of high-grade PIN and 83% of PCA samples. The reduction percentages for each lesion when compared with the normal epithelium were 86% (P<0.001), 72% (P<0.001) and 68% (P<0.01), respectively. TLs of these lesions differed significantly when compared with one another and the normal tissue (P<0.001). These findings, along with AMACR expression, suggest the premalignant nature of AAH and its role in prostate cancer development (71). Lastly, TL in BPH is associated with race, as African American males have been shown to exhibit significantly shorter telomeres compared to Caucasian males; longer telomeres have also been found to be associated with an increased risk of cancer (72).

hTERT promoter mutations (C250T and C228T) are a key event during the carcinogenesis of CNS tumors (73). The frequency of these mutations varies among different tumors, such as primary glioma (80%), medulloblastoma (19.8%) and meningioma (7.4%). The WHO grade of primary gliomas is associated with the frequency of hTERT mutations. Similarly, it has been reported that glioblastomas (grade IV), oligodendrogliomas (grade II-III) and astrocytomas (grade I) present these mutations in 80, 60-70 and 30-40% of cases, respectively. The multi-sector sequencing of glioblastomas has revealed the clonal nature of the hTERT mutations in these tumors, indicating their essential role in the transformation from precancerous lesions to malignant tumors (73). As a result, increased TL and telomerase activation are significant risk factors for glioma and glioblastoma development. In addition, inherited mutations near hTERT and other telomere-related genes, namely hTERC, RTEL1 and POT1, could increase the susceptibility of neural cells to oncogenesis (74).

Actinic keratosis (AK) and Bowen's disease (BD) are pre-invasive, in situ forms of cutaneous SCC (cSCC). Both AK and BD harbor hTERT promoter mutations: -146C>T or C250T, -124C>T or C228T, -138/-139CC>TT or CC242/243TT (genome location: chr.5.1295242_1295243CC>TT) and -124/-125CC>TT or CC228/229TT (genome location: chr.5.1295228_1295229CC>TT), which gradually decrease following treatment (75). Consequently, telomerase activity in those lesions has been found to be increased. Moreover, TL has been shown to be associated with a greater tumor invasiveness, as the telomere centromere ratio values in cases of cSCC are lower than those in BD and AK. It is therefore understood that telomere shortening plays a crucial role in the invasive progression of cSCC from its precursors, as it precedes UV-induced p53 mutations (76).

Benign, precancerous and malignant melanocytic proliferations exhibit different telomerase activity levels, which increase with tumor invasiveness. These findings have been confirmed by immunohistochemistry, as well as by the PCR-based TRAP assay. As expected, benign nevi, such as Spitz and acquired nevi exhibited lower enzyme activity levels than dysplastic nevi, which had similar scores with stage I melanoma (77,78). It should be noted that hTERT promoter mutations were initially described in familial melanoma and subsequently, in sporadic melanoma (79). Notably, a previous study identified hTERT promoter mutations in the early stages of melanoma. A total of 77% of areas of intermediate lesions and melanomas in situ harbored hTERT promoter mutations. This finding indicates that these mutations are selected at an unexpectedly early stage of the neoplastic progression (80).

Of note, telomerase activity has been reported to be higher in other non-malignant conditions, for instance, psoriatic lesioned skin, UV-damaged skin and poison ivy dermatitis (81).

Among head and neck squamous carcinomas, hTERT promoter mutations are frequent in cases which are derived from the oral cavity (82). As regards pre-invasive lesions, immunohistochemical analyses have demonstrated that oral epithelial dysplasia exhibit an increased hTERT expression compared to normal mucosa cells, while in oral SCC, the immunohistochemical expression of the protein has been found to be higher than in the dysplastic and normal tissue (83). Additionally, a sub-type of oral leucoplakia featuring ortho-keratotic dysplasia has been shown to exhibit a shorter TL than SCC in situ and the normal epithelium (84). Oral submucous fibrosis (OSMF) is a potential precursor of oral SCC. A study comparing telomerase activation (hTERT expression) between normal mucosa, OSMF and oral SCC found an increased enzyme activity in the latter two conditions. Finally, hTERT levels increased with the histological grading of the SCCs, which indicates that telomerase reactivation is critical during the malignant transformation of OSMF to SCC (85).

The presence of hotspot hTERT mutations in malignant thyroid tumors has been found to be associated with a worse prognosis and a poor response to treatment (86-89). Apart from hTERT promoter mutations, epigenetic alterations, hTERT gene copy number variations and alternative splicing are implicated in the pathogenesis of thyroid malignancies (90). Several studies have investigated premalignant and benign thyroid nodules as controls for identifying hTERT promoter mutations. The vast majority of the samples have not been found to harbor hTERT mutations (91-97).

A hotspot hTERT promoter C228T mutation was described in a case report of a 68-year-old female with a thyroid follicular adenoma (98). It was considered that hTERT promoter mutations comprised a potential early genetic event in the pathogenesis of follicular thyroid carcinoma (98). Moreover, a study including primary tumors from 58 patients with follicular adenoma, 18 with atypical follicular adenoma with uncertain malignant potential, 52 with follicular carcinoma and 20 negative controls from non-tumorous thyroids lesions revealed hTERT promoter hotspot mutations in one follicular adenoma (C228T), three atypical follicular adenomas (all C228T), nine follicular carcinomas (8 C228T and 1 C250T) and in none of the negative controls (99). The lesions that presented the mutations also tested positive for hTERT mRNA and telomerase activity. The C228T mutation was associated with NRAS gene mutations (P=0.16), the most common mutations in thyroid nodules. The TL was also examined in the follicular adenoma and atypical follicular adenoma specimens; however, no significant difference between the hTERT promoter mutation positive and negative was found (99). Another study examined the frequency of hTERT promoter mutations in 34 well-differentiated thyroid carcinomas, 29 follicular adenomas and 33 sporadic adenomas. hTERT promoter mutations were found in 6 patients with adenoma, although no hTERT promoter mutations were detected in the sporadic adenoma group (100).

As regards hTERT expression, a study found that 12 out of 33 follicular adenomas and 4 out of 31 multinodular goiters were positive for hTERT expression. The difference between them was significant (P=0.03); however, no significant difference was found between follicular adenomas and carcinomas (101). Of note, in another study, a positive hTERT mRNA expression among adenomas was associated with lymphocytic infiltration and thyroiditis rather than a worse prognosis, and it was suggested that since lymphocytes express hTERT, lymphocytic infiltration of the examined tissue may influence hTERT expression analysis (102).

A comprehensive investigation of hTERT-related divergence, namely mRNA expression, promoter mutations, promoter hypermethylation and gene copy number alterations, was explored in a study including 43 follicular adenomas and 33 follicular tumors of uncertain malignant potential (FT-UMP). hTERT mRNA was expressed in 6/43 (14%) adenomas and 9/23 (39%) of FT-UMPs (P=0.020). No hTERT promoter mutations were found in Fas, while 6/32 (19%) FT-UMPs were positive (P=0.005). No difference in median mutation frequency was observed between the FT-UMPs and follicular carcinomas (P=0.858). The promoter methylation intensity was higher in follicular carcinomas (13%) and FT-UMPs (11%) compared to FA (8%) (P<0.001 and P=0.045). hTERT gene copy number exhibited variations (gain or loss) in 5/19 FT-UMPs, similar to 11/77 follicular carcinomas (103).