Abstract
Background
Telomere dysfunction has been reported to be directly involved in carcinogenesis owing to chromosomal instability and immortalization; however, the clinicopathological significance of telomeres remains controversial. We have shown that telomere shortening occurs in normal-appearing duct cells at initiation and then continues during the progression of pancreatic cancer. In this study, we determined the clinicopathological and prognostic value of telomere length (TL) in cancer progression.
Methods
TL in both cancer cells and cancer-associated fibroblasts (CAFs) was analyzed by high-throughput quantitative fluorescence in situ hybridization using a previously reported cohort comprising 1434 cases of adenocarcinoma (ADC), squamous cell carcinoma (SCC), adenosquamous carcinoma, hepatocellular carcinoma, and renal cell carcinoma (RCC), which are known cancers with a statistically significantly low incidence of alternative lengthening of telomeres. Cases were divided into 2 groups as follows: longer and shorter telomeres, according to the median TL of cancer cells and CAFs. The statistical significance of TL in cancer cells and CAFs on clinicopathological characteristics and prognosis was analyzed.
Results
There was a close association between TL in cancer cells and CAFs. Longer telomeres in cancer cells and CAFs were associated with aggressive features such as advanced stage, high mitosis score and nuclear score, poorly differentiated cancer, and desmoplastic stroma in ADC. Furthermore, a longer TL was an independent prognostic factor for ADC, SCC, and RCC.
Conclusions
Longer telomeres are associated with worse prognosis in ADC, SCC, and RCC. Thus, TL is a novel biomarker for the diagnosis of aggressive cancers with poor prognoses.
Immortality, one of the hallmarks of cancer (1), is caused by the telomere length (TL) maintenance mechanism (TMM) (2) because telomeres are shortened with every cell division and finally trigger a DNA damage response that results in senescence and carcinogenesis (3). We previously reported that patients with pancreatic cancers show telomere shortening of normal-appearing duct cells without histological changes using formalin-fixed paraffin-embedded tissue by quantitative fluorescence in situ hybridization (FISH) (4). Telomere attrition of dysplasia and carcinoma in situ, which are precancerous lesions, gradually develops during carcinogenesis. Our previous studies used clinical samples to clarify the alterations in TL at the initiation and promotion stages of carcinogenesis, whereby normal cells are transformed into cancer cells.
Many studies have revealed that TMM is activated in most cancers to counteract telomere shortening and to reduce the number of tumor cells from the onset of telomeric crisis (2,5). Approximately 80%-90% of cancers demonstrate TMM via the enzymatic activity of telomerase (5). A subset of neoplasms, mainly represented by sarcomas, maintain TL via a recombination-based mechanism with a telomerase-independent process called alternative lengthening of telomeres (ALT) (6). Several methods to detect ALT cells exist: C-Circle assay by polymerase chain reaction (PCR) (7), long and heterogeneous telomeres by FISH (8), subnuclear structures termed ALT-associated promyelocytic leukemia bodies by FISH, and immunofluorescent staining for telomeres containing telomeric DNA and telomere-specific binding proteins telomeric repeat-binding factor 1/2 (9). Further, a loss of alpha thalassemia X-linked intellectual disability (ATRX) and death domain associated protein (DAXX) by immunostaining is closely associated with mutations and presence of ALT (10,11).
Epidemiological studies have shown that the relationship between TL and patient prognosis remains controversial (12,13). Meta-analyses have shown that longer TL is related to improved prognosis in leukemia (12), colorectal cancer (14), and esophageal cancer (15), but the association between prognosis and TL in most malignancies is uncertain (13). There might be inconsistencies in the effects of TL on cancer prognosis, possibly due to variable methods and samples such as southern blotting, quantitative PCR, real-time PCR, or quantitative FISH using blood, tumor tissue, adjacent noncancerous tissue, or normal tissue (12-15). Furthermore, many factors such as aging, stress, lifestyle (16), and cancer microenvironment (17,18) influence the TL of blood cells and whole cancer tissue containing cancer cells, stroma, and blood-derived cells. On the other hand, quantitative FISH can accurately measure the TL of each cell separately (19).
Cancer-associated fibroblasts (CAFs) are a key component of the tumor microenvironment with diverse functions (20). Ma et al. (21) reported that cancer cells and CAFs represent telomere attrition compared with their normal counterparts. Telomeres in both cancer cells and CAFs are associated with the prognosis of hepatocellular carcinoma (HCC) (21) and prostate cancer (22); thus, telomeres in CAFs might be a valuable biomarker for some cancers. Although telomeres have been considered an attractive target for cancer therapy, telomere-directed therapeutic efficacy has been limited to cancers with short telomeres (23,24). Therefore, investigation of clinicopathological significance of TL in both cancer cells and CAFs is warranted.
In this study, we assessed the TL of common histological types of cancers from a previously reported cohort (25): adenocarcinoma (ADC), squamous cell carcinoma (SCC), adenosquamous carcinoma (ASC), HCC, and renal cell carcinoma (RCC), which are known cancers with a statistically significantly low incidence of ALT (6,26). The purpose of this study was to investigate the clinicopathological significance, especially the prognostic utility of TL.
Methods
Patients
We used the same cohort as previously reported (25) as follows: ADC (n = 919), SCC (n = 46), ASC (n = 35), HCC (n = 301), RCC (n = 131), and others (neuroendocrine carcinoma of the lung, n = 2). The primary organs were the lungs (n = 340), colon (n = 117), stomach (n = 202), pancreas (n = 251), liver (n = 308), breast (n = 85), and kidneys (n = 131). This study was conducted in accordance with the principles of the Declaration of Helsinki (2013) and was approved by the ethics committees of Kagawa University (approval no. 2019-209), Kanazawa University (approval no. 181 and no. 2016-093), and Kanagawa Cancer Center (approval no. 177).
Pathological assessment
Pathological sections were diagnosed based on the World Health Organization Classification of Tumours (27-30). The pathological stage was determined based on the tumor (T), nodes (N), and metastases (M) (TNM) Classification of Malignant Tumors 7th edition (31).
A board-certified pathologist and authors (Y. Ma., Y.J., K.Y., and K.A.) reviewed the hematoxylin and eosin (H&E)-stained sections for pathological assessments. The gland score and nuclear score were assessed based on the WHO classification of breast tumors (score 1, >75% of tumor cells form glands; score 2, 10%-75% of tumor cells form glands; and score 3, <10% of tumor cells form gland; and score 1, small, regular, uniform cells; score 2, moderate increase in size and variability; and score 3, marked variation) (27). Mitotic counts in 10 high-power fields were scored based on the WHO classification of breast tumors (field diameter: 0.54 mm; score 1, ≤8; score 2, 9-16; and score 3, ≥17) (27). The pathological grade was classified as follows: score 1, well differentiated; score 2, moderately differentiated; score 3, poorly differentiated; and score 4, anaplastic type. Stroma type was based on the Japanese Classification of Colorectal Carcinoma 9th edition, Gastric Carcinoma 14th edition, and Pancreas Carcinoma 7th edition (medullary, abundant cancer cells with small amount of CAFs; scirrhous, few cancer cells with abundant fibrous CAFs; and intermediate, intermediate between medullary and scirrhous). At present, classification of stroma type for lung, liver, breast, and kidney cancers in Japanese Classification of Carcinoma does not exist; thus, we applied the same classification for those cancer types. It was difficult to distinguish lymphocyte and other leukocytes such as multinuclear leukocyte and plasmacyte because tumor infiltrating lymphocytes (32) in some cases of the present cohort were low. Therefore, in this study, we evaluated the degree of infiltrating leukocytes as follows: 0, none; 1, few; 2, moderate; and 3, severe.
Immunohistochemical staining
We performed immunohistochemical staining using rabbit polyclonal anti-ATRX antibody (1:500 dilution, Merck, Burlington, MA, USA) or anti-DAXX antibody (1:500 dilution, Merck, Burlington). To distinguish ADC from SCC and ASC for all lung cancers, immunohistochemical staining was performed for the detection of p40 (prediluted; Nichirei Bioscience Inc., Tokyo, Japan), p63 (prediluted; 7-Jul, Leica Biosystems, Wetzlar, Germany), cytokeratin 5/6 (D5/16B4, prediluted; Agilent Technologies, Inc., Santa Clara, CA, USA), Napsin A (IP64, 1:100 dilution; Leica Biosystems), and TTF1 (8G7G3/1, 1:50 dilution; Agilent Technologies, Inc) (33). Immunostained sections were reviewed by 3 authors (Y. Ma, Y.J., and K.Y.).
High-throughput quantitative FISH for analysis of TL
The slides were processed using FISH as previously reported (4,25). Tissue sections were hybridized with 200 nM peptide nucleic acid probes for the telomere (5-CCCTAACCCTAACCCTAA-3′; Panagene, Daejeon, Korea) and centromere (5-CTTCGTTGGAAACGGGGT-3′; Panagene) for 3 minutes at 80°C followed by 1 hour at 25°C. The nuclei were stained with 4′,6-Diamidine-2′-phenylindole dihydrochloride (DAPI) (Molecular Probes, Eugene, OR, USA).
ImageJ (version 1.53a, Wayne Rasband, National Institutes of Health, modified by the plug-in AsKey, Kagawa, Japan) was used to automatically estimate the fluorescence intensities of the individual nuclei. The nuclear area of each cell was automatically identified by DAPI staining (Supplementary Figure 1, A and B, available online), and then the intensity of red, green, and blue signals in each nuclear area was measured (Supplementary Figure 1, C, available online). To set appropriate background subtraction, we analyzed the intensity of the signals using a human leukemia cell line, 1301, which has long telomeres and a human fibroblast (HFL)-1 (population doubling level [PD] of 20 and 40; Supplementary Figure 1, C-F, available online). The intensity of the top 5% and bottom 40% of red signals showed concordant data with the results of previously known TL (1301>HFL1, PD 20>HFL-1, PD40; Supplementary Figure 1, C, available online); thus, we used the intensity of the top 5% to the bottom 40%. Furthermore, the ratio of red to green signals was better to analyze TL (Supplementary Figure 1, D, available online) compared with the ratio of red to blue signals (Supplementary Figure 1, E, available online). Therefore, we used the ratio of red to green signals (telomere/centromere ratio) for TL assessment (Supplementary Figure 1, F, available online).
HFL-1 (population doubling level of 20) was used as a control, and we performed FISH using HFL-1 (PD20) every time. The normalized telomere signals for each case were calculated as follows: (median value of telomere/centromere ratio of target cells)/(median value of telomere/centromere ratio of control HFL-1 cells). We measured the TL of cancer cells and CAFs separately using the same section (Figure 1, A). To assess CAFs, only spindle cells were analyzed; vessels (lumen forming cells) and leukocytes were excluded from the analysis. The cutoff value of TL was determined by the median value of TL in cancer cells and CAFs because the P value in the log-rank test for overall survival was low at the median value (cancer cells, 0.828; CAFs, 0.885).
Figure 1.
Telomere length (TL) of cancer cells and cancer-associated fibroblasts (CAFs) from 1394 cases represented similar clinicopathological characteristics. (A) Hematoxylin and eosin (H&E) and fluorescence in situ hybridization (FISH) images from a breast adenocarcinoma (ADC) case. Arrowheads, cancer cells; black arrows, fibroblast; yellow arrows, lymphocyte. Red, telomere; green, centromere; blue, 4′,6-Diamidine-2′-phenylindole dihydrochloride (DAPI). (B) TL of cancer cells and CAFs showed a positive correlation. TL of cancer cells (C) and CAFs (D) was not associated with age. Statistical analysis used was Pearson’s coefficient analysis (B-D). Six histological types of cancers showed variety of TL of cancer cells (E) and CAFs (F). Number of cases: ADC, 919; squamous cell carcinoma (SCC), 46; adenosquamous carcinoma (ASC), 35; hepatocellular carcinoma (HCC), 301; renal cell carcinoma (RCC), 131; others (neuroendocrine carcinoma of the lung), 2. Cancers of 7 organs showed variety of TL of cancer cells (G) and CAFs (H). Number of cases: lung, 340; colon, 117; stomach, 202; pancreas, 251; liver, 308; breast, 85; kidney, 131.
Statistical analysis
Differences between groups were compared using the unpaired t test, Mann–Whitney U test, χ2 test, Fisher exact test, or Tukey-Kramer test. Correlations were assessed using Pearson correlation coefficients. Survival was analyzed using Kaplan–Meier curves and the log-rank test. Multivariable analyses were performed for covariates sex, age, TL, and stage. The level of statistical significance was set at P less than .05. Statistical analyses were performed using JMP Pro 14 software (SAS Institute Inc., Cary, NC, USA).
Results
Analysis of TL of cancer cells and CAFs
TLs of cancer cells (white and black arrowheads, Figure 1, A) and CAFs (white and black arrows, Figure 1, A) determined by quantitative FISH showed a positive correlation (Figure 1, B). CAFs showed longer telomeres than cancer cells (median TL, cancer cells: 0.828, CAFs: 0.885, Supplementary Table 1, available online), indicating that telomere attrition in cancer cells is higher than that in CAFs.
Noncancerous tissue usually presents telomere attrition in association with aging (3,4). However, TL of cancer cells (Figure 1, C) and CAFs (Figure 1, D) did not show any correlation with age, suggesting that activation of TMM might influence TL of both cancer cells and CAFs.
Comparison of TL with histological types, high malignancy, and desmoplastic stroma of cancer
The TL of cancer cells and CAFs varied depending on the histological type (Figure 1, E and F; Supplementary Table 1, available online) and primary organs (Figure 1, G and H). ADC, ASC (Supplementary Table 1), breast, stomach, and pancreatic cancers showed statistically significantly longer telomeres in cancer cells and CAFs than in other cancer types.
ADC cases with advanced stage (Figure 2, A and B), high mitosis scores (Figure 2, C and D), and high nuclear scores (Figure 2, E and F) showed longer telomeres in cancer cells and CAFs. A high nuclear score was associated with longer TL in ADCs (Figure 2, E and F), but there was no association between nuclear size (Supplementary Figure 2, A and B, available online) and DAPI intensity (reflecting DNA content; Supplementary Figure 2, B and C, available online) and TL. A high pathological grade of ADC was associated with longer telomeres in cancer cells (Figure 2, G) but not those in CAFs (Figure 2, H), indicating that ADC cases with more malignant pathological characteristics are associated with longer telomeres in cancer cells and CAFs.
Figure 2.
Telomere length (TL) of cancer cells and cancer-associated fibroblasts (CAFs) were associated with aggressive clinicopathological characteristics of cancers. TL of cancer cells and CAFs of adenocarcinoma (ADC) was associated with stage (A and B), mitosis score (C and D), nuclear score (E and F), and stroma type (I and J), but histological grade (G and H) and inflammation score (K and L) did not show any association with TL. TL of cancer cells (A, C, E, G, I, and K) and CAFs (B, D, F, H, J, and L). Tukey-Kramer analysis. ASC = adenosquamous carcinoma; HCC = hepatocellular carcinoma; RCC = renal cell carcinoma; SCC = squamous cell carcinoma.
Surprisingly, scirrhous ADC showed longer telomeres of cancer cells and CAFs than intermediate and medullary ADC (Figure 2, I and J), suggesting that the desmoplastic stroma is involved in the TMM of cancer cells and CAFs. A high inflammation score tended to be associated with longer telomeres in ADCs, but the difference was not statistically significant (Figure 2, K and L).
We have divided breast cancer patients into 3 groups as follows: 1) estrogen receptor (ER)+/HER2−; 2) HER2+; and 3) triple-negative [ER−, progesterone receptor−, and HER2]. However, there was no association between TL and ER, progesterone receptor, and HER2 (Supplementary Figure 2, available online).
Correlation of TL in cancer cells or CAFs with worse prognosis
Patients were divided into 2 groups according to the median TL (Figure 3; Supplementary Figure 3, E-P, available online). In all cancer types, longer telomeres of cancer cells and CAFs were associated with worse overall survival (Figure 3, A and D), and longer telomeres of CAFs were associated with worse disease-free survival (Figure 3, J). All cancer types with longer telomeres were young, female predominant, advanced stage, high nuclear score and histological grade, scirrhous type stroma, and high inflammation score compared with those with shorter telomeres (Supplementary Table 2, available online).
Figure 3.
Longer telomere length (TL) of cancer cells or cancer-associated fibroblasts (CAFs) was associated with worse overall and disease-free survival of cancers. TL of cancer cells (A, B, C, G, H, and I) and CAFs (D, E, F, J, K, and L) of all cancer types (A, D, G, and J), Adenocarcinoma (ADC; B, E, H, and K) and squamous cell carcinoma (SCC; C, F, I, and L). Overall survival (A-F) and disease-free survival (G-L). Cutoff values of TL were based on median value (Supplementary Table 1, available online). P values were calculated by log-rank test. Cases were divided into 2 groups according to the median value of TL in cancer cells and CAFs.
Furthermore, ADCs with longer telomeres of cancer cells and CAFs showed worse overall survival (Figure 3, B and E) and disease-free survival (Figure 3, H and K). In ADC cases with longer telomeres of cancer cells, there were many stomach, pancreatic, and breast cancers, with advanced stage, high nuclear score, high pathological grade, and scirrhous stroma. SCC cases with longer telomeres in cancer cells were predominantly female. These results indicate that the characteristics of cases with longer telomeres differ depending on the histological type of cancer.
In the multivariate analysis, longer telomeres in cancer cells of all cancer types, ADC, and SCC and longer telomeres of CAFs in SCC and RCC were associated with worse overall survival (Table 1). Furthermore, longer telomeres of CAFs in ADC were associated with worse disease-free survival (Table 2). However, ASC and HCC did not show a statistically significant association between TL and prognosis (Supplementary Figure 3, E-P, available online).
Table 1.
Multivariable analysis for OS
| All |
ADC |
SCC |
ASC |
HCC |
RCC |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | |||
| Worse OS related longer telomere in cancer | ||||||||||||||
| Sex, female | 0.80 (0.66 to 0.97) | .02 | 0.77 (0.61 to 0.96) | .02 | 0.38 (0.07 to 2.09) | .26 | 1.01 (0.39 to 2.63) | .98 | 1.31 (0.65 to 2.64) | .46 | 0.61 (0.28 to 1.31) | .20 | ||
| Age, ≥65 y | 1.28 (1.06 to 1.55) | .01 | 1.31 (1.05 to 1.64) | .02 | 1.91 (0.68 to 5.37) | .22 | 0.86 (0.30 to 2.49) | .78 | 0.76 (0.44 to 1.32) | .34 | 1.65 (0.84 to 3.25) | .14 | ||
| Telomere, cancera, long | 1.21 (1.01 to 1.46) | .04 | 1.29 (1.04 to 1.62) | .02 | 4.07 (1.27 to 13.02) | .02 | 1.23 (0.54 to 2.83) | .62 | 1.09 (0.62 to 1.89) | .77 | 0.63 (0.31 to 1.28) | .20 | ||
| TNM stage | ||||||||||||||
| II | 4.61 (3.56 to 5.97) | <.001 | 5.14 (3.57 to 7.42) | <.001 | 0.81 (0.099 to 6.70) | .85 | 1.58 (0.64 to 3.90) | .32 | 3.61 (1.91 to 6.83) | <.001 | 6.63 (1.36 to 32.45) | .02 | ||
| III | 5.05 (3.73 to 6.69) | <.001 | 5.10 (3.40 to 7.65) | <.001 | 25.11 (5.60 to 112.71) | <.001 | 0 (0 to N. D.) | >.99 | 14.19 (6.63 to 30.37) | <.001 | 7.46 (2.99 to 18.60) | <.001 | ||
| IV | 13.32 (8.70 to 20.38) | <.001 | 15.96 (8.52 to 29.89) | <.001 | N. D. | N. D. | N. D. | N. D. | 0 (0 to N. D.) | >.99 | 34.32 (12.92 to 91.16) | <.001 | ||
| Worse OS related longer telomere in CAFs | ||||||||||||||
| Sex, female | 0.81 (0.67 to 0.99) | .04 | 0.78 (0.62 to 0.99) | .04 | 0.46 (0.09 to 2.45) | .36 | 0.90 (0.35 to 2.30) | .82 | 1.31 (0.65 to 2.64) | .46 | 0.76 (0.35 to 1.62) | .46 | ||
| Age, ≥65 y | 1.30 (1.07 to 1.58) | .01 | 1.34 (1.07 to 1.70) | .01 | 1.87 (0.66 to 5.30) | .24 | 0.90 (0.32 to 2.49) | .84 | 0.76 (0.44 to 1.32) | .33 | 1.54 (0.79 to 3.00) | .21 | ||
| Telomere, CAFsa, long | 1.16 (0.95 to 1.40) | .14 | 1.21 (0.96 to 1.53) | .10 | 3.00 (1.01 to 8.90) | .05 | 0.46 (0.18 to 1.19) | .11 | 1.08 (0.61 to 1.93) | .79 | 2.44 (1.18 to 5.04) | .02 | ||
| TNM stage | ||||||||||||||
| II | 4.50 (3.46 to 5.85) | <.001 | 5.09 (3.50 to 7.39) | <.001 | 0.60 (0.07 to 4.88) | .64 | 2.50 (0.91 to 6.85) | .08 | 3.61(1.91 to 6.82) | <.001 | 11.23 (2.21 to 57.18) | .01 | ||
| III | 5.02 (3.73 to 6.76) | <.001 | 5.13 (3.39 to 7.75) | <.001 | 20.81 (4.87 to 88.94) | <.001 | 0 (0 to N. D.) | >.99 | 13.92 (6.50 to 29.80) | <.001 | 6.62 (2.66 to 16.46) | <.001 | ||
| IV | 13.30 (8.68 to 20.37) | <.001 | 15.06 (7.96 to 28.49) | <.001 | N. D. | N. D. | N. D. | N. D. | 0 (0 to N. D.) | >.99 | 38.92 (14.55 to 104.10) | <.001 | ||
| Worse OS related telomere length in cancer/CAFs | ||||||||||||||
| Sex, female | 0.82 (0.67 to 0.96) | .04 | 0.79 (0.63 to 1.00) | .05 | 0.34 (0.06 to 1.97) | .23 | 0.78 (0.30 to 2.03) | .61 | 1.30 (0.64 to 2.63) | .47 | 0.69 (0.32 to 1.48) | .34 | ||
| Age, ≥65 y | 1.32 (1.09 to 1.61) | .004 | 1.39 (1.10 to 1.76) | .01 | 1.92 (0.67 to 5.45) | .22 | 0.88 (0.30 to 2.60) | .82 | 0.76 (0.43 to 1.32) | .33 | 1.64 (0.83 to 3.26) | .15 | ||
| Telomere, cancer/CAFsa | ||||||||||||||
| Short/long | 1.27 (0.93 to 1.73) | .13 | 1.15 (0.79 to 1.67) | .47 | 2.01 (0.20 to 20.03) | .55 | 0.29 (0.06 to 1.38) | .12 | 1.08 (0.43 to 2.75) | .87 | 3.01 (1.04 to 8.74) | .04 | ||
| Long/short | 1.39 (1.03 to 1.88) | .03 | 1.37 (0.95 to 1.96) | .09 | 3.79 (0.61 to 23.74) | .15 | 2.37 (0.57 to 9.77) | .23 | 1.10 (0.44 to 2.73) | .84 | 0.35 (0.10 to 1.62) | .09 | ||
| Long/long | 1.27 (1.02 to 1.59) | .03 | 1.39 (1.07 to 1.82) | .02 | 4.88 (1.32 to 18.00) | .02 | 0.64 (0.23 to 1.77) | .39 | 1.12 (0.58 to 2.13) | .75 | 1.33 (0.56 to 3.18) | .52 | ||
| TNM stage | ||||||||||||||
| II | 4.50 (3.46 to 5.82) | <.001 | 5.08 (3.49 to 7.37) | <.001 | 0.81 (0.093 to 6.45) | .81 | 2.66 (0.96 to 7.38) | .06 | 3.62 (1.91 to 6.86) | <.001 | 10.45 (2.32 to 53.73) | .01 | ||
| III | 5.00 (3.72 to 6.73) | <.001 | 5.21 (3.44 to 7.89) | <.001 | 26.61 (5.74 to 132.72) | <.001 | 0 (0 to N. D.) | >.99 | 14.11 (6.51 to 30.60) | <.001 | 6.85 (2.73 to 17.14) | <.001 | ||
| IV | 14.60 (9.51 to 22.42) | <.001 | 19.96 (10.42 to 38.22) | <.001 | N. D. | N. D. | N. D. | N. D. | 0 (0 to N. D.) | >.99 | 58.72 (19.90 to 173.28) | <.001 | ||
The cutoff value was median telomere length (TL; cancer cells: 0.828, cancer-associated fibroblast [CAFs]: 0.885). TNM stage, UICC 7th edition. ADC = adenocarcinoma; ASC = adenosquamous carcinoma; HCC = hepatocellular carcinoma; HR = hazard ratios; N. D. = not determined; OS = overall survival; RCC = renal cell carcinoma; SCC = squamous cell carcinoma.
Table 2.
Multivariable analysis for DFS
| All |
ADC |
SCC |
ASC |
HCC |
RCC |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | |
| Worse DFS related longer telomere in cancer | ||||||||||||
| Sex, female | 0.90 (0.76 to 1.06) | .22 | 1.00 (0.82 to 1.24) | .95 | 0.342 (0.04 to 2.89) | .32 | 1.11 (0.43 to 2.84) | .83 | 1.21 (0.82 to 1.80) | .34 | 1.13 (0.43 to 2.95) | .80 |
| Age, ≥65 y | 1.19 (1.01 to 1.41) | .04 | 1.07 (0.87 to 1.32) | .51 | 1.44 (0.51 to 4.09) | .50 | 0.82 (0.29 to 2.31) | .70 | 1.10 (0.78 to 1.55) | .58 | 0.81 (0.30 to 2.16) | .68 |
| Telomere, cancera, long | 0.95 (0.80 to 1.12) | .53 | 1.16 (0.94 to 1.42) | .17 | 2.84 (0.93 to 8.70) | .07 | 1.54 (0.58 to 4.13) | .39 | 0.98 (0.71 to 1.36) | .91 | 1.00 (0.39 to 2.59) | .99 |
| TNM stage | ||||||||||||
| II | 3.29 (2.69 to 4.02) | <.001 | 5.61 (3.95 to 7.96) | <.001 | 1.63 (0.34 to 7.80) | .54 | 1.87 (0.56 to 6.26) | .31 | 1.89 (1.33 to 2.65) | <.001 | 13.08 (2.36 to 72.64) | .003 |
| III | 3.27 (2.56 to 4.18) | <.001 | 5.72 (3.89 to 8.42) | <.001 | 11.87 (2.40 to 58.84) | .002 | 0 (0 to N. D.) | >.99 | 4.77 (2.65 to 8.59) | <.001 | 14.32 (4.67 to 43.92) | <.001 |
| IV | 14.74 (9.26 to 23.47) | <.001 | 22.89 (12.81 to 40.90) | <.001 | N. D. | N. D. | N. D. | N. D. | 9.57 (2.91 to 31.42) | <.001 | 607.40 (29.75 to 12399.59) | <.001 |
| Worse DFS related longer telomere in CAFs | ||||||||||||
| Sex, female | 0.87(0.73 to 1.03) | .10 | 0.96 (0.77 to 1.19) | .71 | 0.37 (0.04 to 3.12) | .36 | 0.87 (0.34 to 2.19) | .77 | 1.21 (0.82 to 1.80) | .34 | 1.12 (0.44 to 2.85) | .81 |
| Age, ≥65 y | 1.22 (1.03 to 1.45) | .02 | 1.12 (0.90 to 1.39) | .30 | 1.33 (0.46 to 3.80) | .60 | 0.82 (0.29 to 2.27) | .69 | 1.10 (0.78 to 1.55) | .58 | 0.79 (0.30 to 2.09) | .65 |
| Telomere, CAFsa, long | 1.01 (0.85 to 1.19) | .95 | 1.27 (1.02 to 1.57) | .03 | 2.61 (0.85 to 8.01) | .09 | 0.80 (0.28 to 2.29) | .68 | 0.96 (0.69 to 1.33) | .82 | 1.32 (0.52 to 3.36) | .56 |
| TNM stage | ||||||||||||
| II | 3.24 (2.64 to 3.98) | <.001 | 5.48 (3.84 to 7.82) | <.001 | 1.53 (0.32 to 7.34) | .60 | 2.66 (0.71 to 9.88) | .14 | 1.88 (1.33 to 2.64) | <.001 | 14.22 (2.52 to 80.32) | .003 |
| III | 3.26 (2.54 to 4.17) | <.001 | 5.59 (3.77 to 8.29) | <.001 | 11.26 (2.27 to 55.78) | .003 | 0 (0 to N. D.) | >.99 | 4.82 (2.67 to 8.68) | <.001 | 13.88 (4.53 to 42.48) | <.001 |
| IV | 14.70 (9.20 to 23.48) | <.001 | 21.53 (11.93 to 38.83) | <.001 | N. D. | N. D. | N. D. | N. D. | 9.51 (2.90 to 31.20) | <.001 | 540.96 (26.33 to 11112.44) | <.001 |
| Worse DFS related telomere length in cancer/CAFs | ||||||||||||
| Sex, female | 0.87 (0.73 to 1.04) | .13 | 1.04 (0.84 to 1.30) | .69 | 0.357 (0.04 to 3.08) | .35 | 1.03 (0.40 to 2.68) | .95 | 1.20 (0.81 to 1.78) | .37 | 1.09 (0.42 to 2.86) | .86 |
| Age, ≥65 | 1.22 (1.02 to 1.44) | .02 | 1.12 (0.90 to 1.38) | .33 | 1.30 (0.44 to 3.78) | .64 | 0.75 (0.61 to 2.27) | .61 | 1.08 (0.77 to 1.53) | .65 | 0.87 (0.32 to 2.39) | .79 |
| Telomere, cancer/CAFsa | ||||||||||||
| Short/long | 1.10 (0.84 to 1.45) | .49 | 1.46 (1.04 to 2.04) | .03 | 4.72 (0.71 to 31.26) | .11 | 0.74 (0.12 to 4.43) | .74 | 1.18 (0.71 to 1.96) | .52 | 1.96 (0.47 to 8.12) | .35 |
| Long/short | 0.99 (0.75 to 1.32) | .95 | 1.29 (0.91 to 1.82) | .15 | 6.25 (0.94 to 41.69) | .06 | 2.58 (0.56 to 11.81) | .22 | 1.22 (0.75 to 1.99) | .43 | 1.29 (0.29 to 5.74) | .74 |
| Long/long | 0.97 (0.80 to 1.18) | .79 | 1.32 (1.02 to 1.69) | .03 | 3.97 (1.00 to 15.80) | .05 | 1.14 (0.34 to 3.84) | .83 | 0.95 (0.65 to 1.39) | .78 | 1.29 (0.39 to 4.23) | .67 |
| TNM stage | ||||||||||||
| II | 3.25 (2.65 to 3.99) | <.001 | 5.54 (3.87 to 7.92) | <.001 | 1.82 (0.36 to 9.21) | .46 | 2.43 (0.63 to 9.31) | .20 | 1.89 (1.34 to 2.66) | <.001 | 14.81 (2.59 to 84.85) | .003 |
| III | 3.25 (2.54 to 4.17) | <.001 | 5.66 (3.81 to 8.39) | <.001 | 17.59 (3.09 to 100.29) | .001 | 0 (0 to N. D.) | >.99 | 4.73 (2.60 to 8.62) | <.001 | 13.29 (4.28 to 41.22) | <.001 |
| IV | 14.61 (9.14 to 23.34) | <.001 | 20.95 (11.60 to 37.85) | <.001 | N. D. | N. D. | N. D. | N. D. | 10.07 (3.05 to 33.26) | <.001 | 556.00 (29.97 to 11464.24) | <.001 |
The cutoff value was median telomere length (TL; cancer cells: 0.828, cancer-associated fibroblasts [CAFs]: 0.885). TNM stage, UICC 7th edition. ADC = adenocarcinoma; ASC = adenosquamous carcinoma; DFS = disease-free survival; HCC = hepatocellular carcinoma; HR = hazard ratios; N. D. = not determined; RCC = renal cell carcinoma; SCC = squamous cell carcinoma.
TL characteristics of both cancer cells and CAFs in worse prognosis
We divided patients into 4 groups according to each median value of TL of cancer cells/CAFs as follows: 1) short/short, 2) short/long, 3) long/short, and 4) long/long. Overall survival of long/long groups was worse compared with short/short groups in all histological types of cancers (Figure 4, A-F). In multivariate analysis, long/long groups in ADC and SCC and the short/long group in RCC were associated with worse overall survival (Table 1). Short/long and long/long groups in ADC were associated with worse disease-free survival (Table 2). These data indicate that long telomeres of cancer cells and CAFs are closely associated with poorer prognosis, independent of TNM stage, supporting its prognostic value in cancers. Furthermore, long/long groups of all cancer types, ADC, and SCC showed similar clinicopathological characteristics to those of longer telomere groups divided by median value (Supplementary Table 3, available online). In addition, the long/long group in all cancer types showed higher inflammation scores than those in short/short group (Supplementary Table 3, available online).
Figure 4.
Longer telomere length (TL) of both cancer cells and cancer-associated fibroblasts (CAFs) was associated with worse overall and disease-free survival of cancers. Overall survival (A-F) and disease-free survival (G-L) of all cancer types (A and G), adenocarcinoma (ADC; B and H), squamous cell carcinoma (SCC; C and I), adenosquamous carcinoma (ASC; D and I), hepatocellular carcinoma (HCC; E and J), and renal cell carcinoma (RCC; F and K). Patients were divided into 4 groups according to TL of cancer cells/CAFs as follows: short/short, short/long, long/short, and long/long. The cutoff value of TL was the median value of TL in cancer cells and CAFs. P values were calculated by log-rank test.
Comparison of top 50 and bottom 50 and 10 cases of TL
We selected the top 50 cases with long telomeres and bottom 50 cases with short telomeres according to the TL of the cancer cells and CAFs (Supplementary Table 4, available online) to clarify the characteristics of cases with very long or short telomeres. Loss of ATRX and DAXX by immunostaining was closely related to their mutations and the presence of ALT (10,11); however, all top 50 cases with long telomeres expressed ATRX and DAXX (Supplementary Figure 4, A and B, available online), indicating that they had no mutations. Furthermore, no cases showed marked enhancement or heterogeneity of telomere signals (Supplementary Figure 4, C and D, available online), which is a marker of ALT by FISH (8). These results indicate that TMM in the top 50 cases was not related to ALT.
The top 50 patients of cancer cell TL in all cancer types were young (Supplementary Table 4, available online), with predominantly stomach, pancreas, and breast cancers; predominantly ADC; advanced stage; high nuclear score; high mitosis score; and predominantly scirrhous type stroma (Supplementary Figure 4, E, available online). In contrast, the bottom 50 patients in all cancer types were associated with predominantly lung cancers, SCC, and HCC; early stage; low nuclear score; low mitosis score; and predominantly medullary-type stroma (Supplementary Figure 4, F, available online).
Discussion
This study revealed the following findings: 1) the TLs of cancer cells and CAFs were closely associated with each other; 2) ADC cases with longer telomeres showed aggressive phenotypes such as advanced stage, high mitosis score, high nuclear score, poorly differentiated cancer, and desmoplastic stroma; 3) TL of cancer cells and/or CAFs was an independent risk factor for ADC, SCC, and RCC. In this study, we also analyzed TL in each organ; however, there was no relationship between TL in each organ and prognosis (data not shown), indicating that difference in histologic type may be important for TMM.
Previous studies have shown that transforming growth factor β (TGF-β) induces telomere dysfunction by TERRA inhibition, a long noncoding RNA with the same sequence as telomeres (17). TGF-β induces telomere dysfunction by regulating Snail1 and telomeric repeat-binding factor 1/2 and, consequently, telomere shortening (34). CAFs are not neoplastic; however, our findings demonstrated that they possess similar characteristics to cancer cells in terms of TL, indicating that CAFs might be influenced by cancer cells. In this study, the type of stroma and degree of infiltrating inflammatory cells were associated with TL; therefore, there is a possibility that epithelial-mesenchymal transition (EMT)-related molecules, including TGF-β, derived from the cancer microenvironment might contribute to TL in both cancer cells and CAFs. Furthermore, human telomerase reverse transcriptase (hTERT) can accelerate EMT (34); thus, reactivation of hTERT in cancer cell expression may be involved in both TMM and EMT.
The high proliferation of cases with longer telomeres might reflect the prevention of telomere crisis by TMM activation. High nuclear atypia in cases with longer telomeres may reflect the progression of chromosomal instability, suggesting telomeric dysfunction in cases with longer telomeres. The association of TL with desmoplastic stroma and infiltrating inflammatory cells might be associated with EMT-related cytokines, such as TGF-β, mainly from fibroblasts (17) and oxidative stress from leukocytes (18). It is unclear why advanced and poorly differentiated cancers harbor longer telomeres, but there might be several hypotheses as follows: 1) TMM activity is highly associated with cancer progression at the late stage of carcinogenesis; or 2) cases with longer telomeres are prone to aggressive cancers.
Using the same cohort as in this study, we performed immunostaining for phosphorylated hTERT, which can determine RNA-dependent RNA polymerase activity, a nontelomeric function of hTERT (25,35). Although we did not find a statistically significant association between phosphorylated hTERT activity and TL in our cohort (data not shown), cases with highly phosphorylated hTERT activity represented similar clinicopathological characteristics to cases with longer telomeres revealed by this study.
Southern blotting is considered as the gold standard for measuring TL; however, it cannot measure TLs in heterogeneous cell populations. In contrast, FISH using a peptide nucleic acid probe on tissue sections can measure the TL of each cell type present in a tissue (36). We used fluorescent intensity of centromeres as the internal control for telomere measurement and calibrated using a block of cultured cells of known TL on the same slide (4). The TL in each cell of a tissue or culture was distributed over an extremely wide range; thus, we analyzed more than 100 cells of each cell type using our original plug-in tool on ImageJ, which allowed automatic identification of nuclear regions and measurements of the telomere/centromere fluorescent ratio. Flow cytometry determined that our quantitative FISH was highly sensitive (37,38). Furthermore, our new image analysis method can analyze TL with high throughput; thus, telomere measurements can be performed using pathology specimens.
This study has several limitations. CAFs were discriminated based on morphology. The numbers of SCC and ASC cases were small. We did not analyze mutations in telomere- or telomerase-related molecules. Next, we will conduct a similar study using ALT cases to clarify the clinicopathological significance of TMM in cancer.
In conclusion, the TL of cancer cells and CAFs is strongly related to each other through the influence of the TMM and the cancer microenvironment. Furthermore, our findings provide a basis for the development of a novel clinical diagnostic tool to identify patients with aggressive cancer and clearly demonstrate the effectiveness of targeted therapy for TMM.
Supplementary Material
Contributor Information
Yoko Matsuda, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Juanjuan Ye, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Keiko Yamakawa, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Yuri Mukai, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Kazuki Azuma, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Linxuan Wu, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan; Department of Plastic Surgery, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Kenkichi Masutomi, Division of Cancer Stem Cell, National Cancer Center Research Institute, Chuo-ku, Tokyo, Japan.
Taro Yamashita, Department of Gastroenterology, Kanazawa University Graduate School of Medical Sciences, Kanazawa, Ishikawa, Japan.
Yataro Daigo, Department of Medical Oncology and Cancer Center; Center for Advanced Medicine Against Cancer, Shiga University of Medical Science, Otsu, Shiga, Japan; Center for Antibody and Vaccine Therapy, Research Hospital, Institute of Medical Science Hospital, The University of Tokyo, Tokyo, Japan.
Yohei Miyagi, Kanagawa Cancer Center Research Institute, Asahi-ku, Yokohama, Japan.
Tomoyuki Yokose, Department of Pathology, Kanagawa Cancer Center, Asahi-ku, Yokohama, Japan.
Takashi Oshima, Department of Gastrointestinal Surgery, Kanagawa Cancer Center, Asahi-ku, Yokohama, Japan.
Hiroyuki Ito, Department of Thoracic Surgery, Kanagawa Cancer Center, Asahi-ku, Yokohama, Japan.
Soichiro Morinaga, Department of Hepato-Biliary and Pancreatic Surgery, Kanagawa Cancer Center, Asahi-ku, Yokohama, Japan.
Takeshi Kishida, Department of Urology, Kanagawa Cancer Center, Asahi-ku, Yokohama, Japan.
Toshinari Minamoto, Divison of Translational and Clinical Oncology, Cancer Research Institute, Kanazawa University, Kanazawa, Japan.
Motohiro Kojima, Division of Pathology, Exploratory Oncology Research and Clinical Trial Center, National Cancer Center, Kashiwa-shi, Chiba, Japan.
Shuichi Kaneko, Department of Gastroenterology, Kanazawa University Graduate School of Medical Sciences, Kanazawa, Ishikawa, Japan.
Reiji Haba, Diagnostic Pathology, Kagawa University, Kita-gun, Kagawa, Japan.
Keiichi Kontani, Department of Thoracic, Breast and Endocrine Surgery, Kagawa University, Kita-gun, Kagawa, Japan.
Nobuhiro Kanaji, Department of Internal Medicine, Division of Hematology, Rheumatology and Respiratory Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Keiichi Okano, Department of Gastroenterological Surgery, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Mariko Muto-Ishizuka, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Masanao Yokohira, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Kousuke Saoo, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Katsumi Imaida, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Futoshi Suizu, Oncology Pathology, Department of Pathology and Host-Defense, Faculty of Medicine, Kagawa University, Kita-gun, Kagawa, Japan.
Funding
This work was supported in part by a grant-in-aid from AMED (Grant Number 20cm0106473h0001 to Y. Ma.), an Extramural Collaborative Research Grant of the Cancer Research Institute, Kanazawa University to Y. Ma. and F. S., a Taiju Life Social Welfare Foundation to Y. Ma. and F. S., and Grant-in-Aid for Scientific Research on Innovative Areas from the Japan Society for the Promotion of Science (JSPS) KAKENHI (Grant Number JP: 16H06277 to Y. M.; 19H03447 to F. S.).
Notes
Role of the funder: The funders did not play a role in any of the following: the design of the study, data collection, data analysis, data interpretation, manuscript writing, and decision to submit the manuscript for publication.
Disclosures: None of the authors has relevant relationships to disclose.
Author contributions: Data curation, J. Y., K. Y., K. A. and Y. Ma.; Funding acquisition, Y. Ma.; Investigation, J. Y., Y. Mu., K. Y., K. A., L. W., M. M-I.,Y. Ma., and F. S.; Methodology, Y. Ma.; Resources, T. Y., Y. D., Y. Mi., T. O., H. I., S. M., T. K., T. M., M. K., S. K., R. H., K. K., N. K., and K. O.; Supervision, K. M., M. Y., K. I., and F. S.; Writing—original draft, Y. Ma.; Writing—review and editing, Y. Ma, T. M., and F. S.
Acknowledgements: We would like to acknowledge the significant contribution of our coauthor, Yoko Matsuda, who was the supervising author for this study. Prof. Matsuda died September 17, 2022. We thank Ms Misa Tanimoto and Ms Masumi Ito (Oncology Pathology, Kagawa University) for their technical support and Ms Maiko Tada and Ms Etsuyo Matsubara (Oncology Pathology, Kagawa University) for their assistance in preparing the manuscript.
Disclaimers: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Data availability
The data are publicly available and can be downloaded here: https://doi.org/10.6084/m9.figshare.20443215.
References
- 1. Hanahan D, Weinberg RA.. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646-674. [DOI] [PubMed] [Google Scholar]
- 2. Nersisyan L, Hopp L, Loeffler-Wirth H, et al. Telomere length maintenance and its transcriptional regulation in Lynch syndrome and sporadic colorectal carcinoma. Front Oncol. 2019;9:1172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Martinez P, Blasco MA.. Replicating through telomeres: a means to an end. Trends Biochem Sci. 2015;40(9):504-515. [DOI] [PubMed] [Google Scholar]
- 4. Matsuda Y, Ishiwata T, Izumiyama-Shimomura N, et al. Gradual telomere shortening and increasing chromosomal instability among PanIN grades and normal ductal epithelia with and without cancer in the pancreas. PLoS One. 2015;10(2):e0117575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Shay JW, Wright WE.. Role of telomeres and telomerase in cancer. Semin Cancer Biol. 2011;21(6):349-353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Heaphy CM, Subhawong AP, Hong SM, et al. Prevalence of the alternative lengthening of telomeres telomere maintenance mechanism in human cancer subtypes. Am J Pathol. 2011;179(4):1608-1615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Henson JD, Lau LM, Koch S, et al. The C-Circle assay for alternative-lengthening-of-telomeres activity. Methods. 2017;114:74-84. [DOI] [PubMed] [Google Scholar]
- 8. Kunieda J, Yamashita K, Togashi Y, et al. High prevalence of TERT aberrations in myxoid liposarcoma: TERT reactivation may play a crucial role in tumorigenesis. Cancer Sci. 2022;113(3):1078-1089. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Venturini L, Erdas R, Costa A, et al. ALT-associated promyelocytic leukaemia body (APB) detection as a reproducible tool to assess alternative lengthening of telomere stability in liposarcomas. J Pathol. 2008;214(4):410-414. [DOI] [PubMed] [Google Scholar]
- 10. De Vitis M, Berardinelli F, Sgura A.. Telomere length maintenance in cancer: at the crossroad between telomerase and alternative lengthening of telomeres (ALT). IJMS. 2018;19(2):606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Singhi AD, Liu TC, Roncaioli JL, et al. Alternative lengthening of telomeres and loss of DAXX/ATRX expression predicts metastatic disease and poor survival in patients with pancreatic neuroendocrine tumors. Clin Cancer Res. 2017;23(2):600-609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Adam R, Diez-Gonzalez L, Ocana A, et al. Prognostic role of telomere length in malignancies: a meta-analysis and meta-regression. Exp Mol Pathol. 2017;102(3):455-474. [DOI] [PubMed] [Google Scholar]
- 13. Smith L, Luchini C, Demurtas J, et al. Telomere length and health outcomes: an umbrella review of systematic reviews and meta-analyses of observational studies. Ageing Res Rev. 2019;51:1-10. [DOI] [PubMed] [Google Scholar]
- 14. Jia H, Wang Z.. Telomere length as a prognostic factor for overall survival in colorectal cancer patients. Cell Physiol Biochem. 2016;38(1):122-128. [DOI] [PubMed] [Google Scholar]
- 15. Zhang C, Chen X, Li L, et al. The association between telomere length and cancer prognosis: evidence from a meta-analysis. PLoS One. 2015;10(7):e0133174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Mehrsafar AH, Serrano Rosa MA, Moghadam Zadeh A, et al. Stress, professional lifestyle, and telomere biology in elite athletes: a growing trend in psychophysiology of sport. Front Psychol. 2020;11:567214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Sung JY, Cheong JH.. Alternative lengthening of telomeres is mechanistically linked to potential therapeutic vulnerability in the stem-like subtype of gastric cancer. Clin Transl Med. 2021;11(9):e561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Reichert S, Stier A.. Does oxidative stress shorten telomeres in vivo? A review. Biol Lett. 2017;13(12):20170463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Aida J, Izumiyama-Shimomura N, Nakamura K, et al. Basal cells have longest telomeres measured by tissue Q-FISH method in lingual epithelium. Exp Gerontol. 2008;43(9):833-839. [DOI] [PubMed] [Google Scholar]
- 20. Sahai E, Astsaturov I, Cukierman E, et al. A framework for advancing our understanding of cancer-associated fibroblasts. Nat Rev Cancer. 2020;20(3):174-186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Ma LJ, Wang XY, Duan M, et al. Telomere length variation in tumor cells and cancer-associated fibroblasts: potential biomarker for hepatocellular carcinoma. J Pathol. 2017;243(4):407-417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Heaphy CM, Yoon GS, Peskoe SB, et al. Prostate cancer cell telomere length variability and stromal cell telomere length as prognostic markers for metastasis and death. Cancer Discov. 2013;3(10):1130-1141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Thompson CAH, Wong JMY.. Non-canonical functions of telomerase reverse transcriptase: emerging roles and biological relevance. Curr Top Med Chem. 2020;20(6):498-507. [DOI] [PubMed] [Google Scholar]
- 24. Ivancich M, Schrank Z, Wojdyla L, et al. Treating cancer by targeting telomeres and Telomerase. Antioxidants (Basel). 2017;6 (1):15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Matsuda Y, Yamashita T, Ye J, et al. Phosphorylation of hTERT at threonine 249 is a novel tumor biomarker of aggressive cancer with poor prognosis in multiple organs. J Pathol. 2022;257(2):172-185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Lawlor RT, Veronese N, Pea A, et al. Alternative lengthening of telomeres (ALT) influences survival in soft tissue sarcomas: a systematic review with meta-analysis. BMC Cancer. 2019;19(1):232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. WHO. Classification of Tumours. Breast Tumours. 5th ed. WHO Classification of Tumours Editorial Board, ed. World Health Organization; 2019. [Google Scholar]
- 28. WHO. Classification of Tumours of the Urinary System and Male Genital Organs. 4th ed. WHO Classification of Tumours Editorial Board, ed. World Health Organization; 2016. [Google Scholar]
- 29. WHO. Classification of Tumours of the Lung, Pleura, Thymus and Heart. 4th ed. WHO Classification of Tumours Editorial Board, ed. World Health Organization; 2015. [Google Scholar]
- 30. WHO. Classification of Tumours, Digestive System Tumours. 5th ed. WHO Classification of Tumours Editorial Board, ed. World Health Organization; 2019. [Google Scholar]
- 31. Sobin LH. TNM Classification of Malignant Tumours. 7th ed. Wiley Blackwell; 2010. [Google Scholar]
- 32. Salgado R, Denkert C, Demaria S, et al. International TILs Working Group 2014. The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: recommendations by an International TILs Working Group 2014. Ann Oncol. 2015;26(2):259-271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. The Japan Lung Cancer Society . General Rule for Clinical and Pathological Record of Lung Cancer. 8th ed. Tokyo, Japan: Kanehara Shuppan; 2016. [Google Scholar]
- 34. Imran SAM, Yazid MD, Idrus RBH, et al. Is there an interconnection between epithelial-mesenchymal transition (EMT) and telomere shortening in aging? IJMS. 2021;22(8):3888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Yasukawa M, Ando Y, Yamashita T, et al. CDK1 dependent phosphorylation of hTERT contributes to cancer progression. Nat Commun. 2020;11(1):1557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Lansdorp PM, Verwoerd NP, van de Rijke FM, et al. Heterogeneity in telomere length of human chromosomes. Hum Mol Genet. 1996;5(5):685-691. [DOI] [PubMed] [Google Scholar]
- 37. Takubo K, Aida J, Izumiyama-Shimomura N, et al. Changes of telomere length with aging. Geriatr Gerontol Int. 2010;10(suppl 1):S197-S206. [DOI] [PubMed] [Google Scholar]
- 38. Aida J, Izumiyama-Shimomura N, Nakamura K, et al. Determination of telomere length by the quantitative fluorescence in situ hybridization (Q-FISH) method. Am J Anal Chem. 2014;05(12):775-783. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data are publicly available and can be downloaded here: https://doi.org/10.6084/m9.figshare.20443215.