TERT promoter mutations in bladder cancer affect patient ... - PNAS

TERT promoter mutations in bladder cancer affect patient survival and disease recurrence through modification by a common polymorphism P. Sivaramakrishna Rachakondaa, Ismail Hosena, Petra J. de Verdierb,c, Mahdi Fallaha, Barbara Heidenreicha, Charlotta Rykb,c, N. Peter Wiklundb,c, Gunnar Steineckd,e, Dirk Schadendorff, Kari Hemminkia,g, and Rajiv Kumara,1 a Division of Molecular Genetic Epidemiology, German Cancer Research Center, 69120 Heidelberg, Germany; bDepartment of Urology, Karolinska University Hospital, 171 76 Stockholm, Sweden; cUrology Laboratory, Department of Molecular Medicine and Surgery, Karolinska Institutet, 171 76 Stockholm, Sweden; d Division of Clinical Cancer Epidemiology, Sahlgrenska Academy, University of Gothenburg, 413 45 Gothenburg, Sweden; eDepartment of Clinical Cancer Epidemiology, Karolinska Intstitutet, 171 76 Stockholm, Sweden; f Department of Dermatology, University Hospital Essen, 45122 Essen, Germany; and g Center for Primary Health Care Research, Lund University, 221 00 Malmö, Sweden

The telomerase reverse transcriptase (TERT) promoter, an important element of telomerase expression, has emerged as a target of cancer-specific mutations. Originally described in melanoma, the mutations in TERT promoter have been shown to be common in certain other tumor types that include glioblastoma, hepatocellular carcinoma, and bladder cancer. To fully define the occurrence and effect of the TERT promoter mutations, we investigated tumors from a well-characterized series of 327 patients with urothelial cell carcinoma of bladder. The somatic mutations, mainly at positions −124 and −146 bp from ATG start site that create binding motifs for E-twenty six/ternary complex factors (Ets/TCF), affected 65.4% of the tumors, with even distribution across different stages and grades. Our data showed that a common polymorphism rs2853669, within a preexisting Ets2 binding site in the TERT promoter, acts as a modifier of the effect of the mutations on survival and tumor recurrence. The patients with the mutations showed poor survival in the absence [hazard ratio (HR) 2.19, 95% confidence interval (CI) 1.02–4.70] but not in the presence (HR 0.42, 95% CI 0.18–1.01) of the variant allele of the polymorphism. The mutations in the absence of the variant allele were highly associated with the disease recurrence in patients with Tis, Ta, and T1 tumors (HR 1.85, 95% CI 1.11–3.08). The TERT promoter mutations are the most common somatic lesions in bladder cancer with clinical implications. The association of the mutations with patient survival and disease recurrence, subject to modification by a common polymorphism, can be a unique putative marker with individualized prognostic potential.

|

|

TERT mutations cancer genetics transcription factors noncoding mutations reporter assays

|

for E-twenty six (Ets) transcription factors and a general binding CCGGAA/T motif for ternary complex factors (TCF) (8). Simultaneous screening of the TERT promoter for somatic mutations in melanoma tumors by us and others resulted in detection of highly recurrent and mutually exclusive somatic mutations mainly at two residues at −124 (1295228) and −146 (1295250) bp from the ATG start site in the TERT promoter (7, 9, 10). The C > T (G > A) transition at both sites also resulted in creation of the Ets/TCF binding motifs. A concurrent study on cancer cell lines indicated the possibility of mutations being present in cancers other than melanoma (10). The occurrence of the TERT promoter mutations in a range of tumors in certain cancer types has now been confirmed (11). In this study, we investigated a large series of 327 patients with well-characterized tumors from urothelial cell carcinoma of bladder for mutations in the TERT promoter. We observed that the common TERT promoter mutations affected two-thirds of the tumors and were spread across all stages and grades, suggesting that the mutations are early events in bladder carcinogenesis. We also assessed the association of those recurrent TERT promoter mutations with patient survival and disease recurrence. Significance This study shows that the telomerase reverse transcriptase (TERT) promoter mutations, which create de novo E-twenty six/ ternary complex factors (Ets/TCF) transcription binding sites, besides being the most common somatic genetic lesions, influence both survival and disease recurrence in bladder cancer patients. The effect of the TERT promoter mutations on both survival and recurrence is modified by a common polymorphism within the preexisting Ets binding site in the TERT promoter. The data were supported by the results from reporter assays carried out in two urothelial carcinoma cell lines. The findings of the study suggest that the TERT promoter mutations in conjunction with the common polymorphism have potential of being used as clinical biomarkers in bladder cancer.

|

he human telomerase reverse transcriptase (TERT) gene encodes the catalytic reverse transcriptase subunit of telomerase that maintains telomere length (1). Increased telomerase activity is perceived to be one of the hallmarks of human cancers, and the transcriptional regulation of the TERT gene is the major cause of its cancer-specific activation (2, 3). The TERT promoter has been shown to be the most important element of telomerase expression because it harbors binding sites for numerous cellular transcriptional activators and repressors that directly or indirectly regulate the gene expression (4). In addition, the high GC content around the transcription start site of the TERT promoter confers epigenetic regulation through methylation and chromatin remodeling (5, 6). We previously described a high-penetrant disease segregating single-nucleotide germ-line mutation in the TERT promoter in a large melanoma family (7). The A > C (T > G) mutation at position −57 bp (from the ATG start site; Chr 5:1295161 hg19 coordinate) resulted in creation of a new binding motif GGAA/T

T

www.pnas.org/cgi/doi/10.1073/pnas.1310522110

Author contributions: P.S.R., D.S., K.H., and R.K. designed research; P.S.R., I.H., P.J.d.V., B.H., and R.K. performed research; P.S.R., P.J.d.V., M.F., C.R., N.P.W., G.S., D.S., K.H., and R.K. contributed new reagents/analytic tools; P.S.R., I.H., P.J.d.V., M.F., B.H., C.R., N.P.W., G.S., D.S., K.H., and R.K. analyzed data; and P.S.R., I.H., P.J.d.V., M.F., B.H., C.R., N.P.W., G.S., D.S., K.H., and R.K. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. E.L.L.D. is a guest editor invited by the Editorial Board. 1

To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1310522110/-/DCSupplemental.

PNAS Early Edition | 1 of 6

GENETICS

Edited by Eros L. Lazzerini Denchi, The Scripps Research Institute, La Jolla, CA, and accepted by the Editorial Board September 5, 2013 (received for review June 3, 2013)

Results We observed that two nucleotide positions at −124 and −146 bp from ATG start site accounted for over 99% of all of the detected mutations in TERT promoter. The −124C > T (G > A) mutation (Fig. S1) was the most frequent alteration that was detected in 175 (53.5%) tumors. In three tumors the nucleotide change involved a C > A (G > T) tranversion at the −124 position. An additional 38 (11.6%) tumors carried the −146C > T (G > A) mutation (Fig. S1). One bladder tumor carried a −57A > C (T > G) mutation (Fig. S1), previously detected as a causal mutation in a melanoma pedigree (7). The mutations at −57, −124, and −146 bp were mutually exclusive and resulted in creation of a general CCGGAA/T binding motif for Ets/TCF transcription factors (Fig. S2). Four tumors, in addition to two recurrent mutations, also carried other changes within the amplified region of the TERT promoter (Table S1). One tumor (D:078) in addition to the −146C > T (G > A) mutation had on the same allele a single nucleotide (G/C) deletion at −129 bp (1295233; Table S1 and Fig. S1). The deletion resulted in depletion of a Sp1 transcription factor binding site. Sequencing of the DNA from corresponding germ-line DNA confirmed that mutations in tumors were somatic (more details are provided in SI Results). Of 327 tumors, 214 (65.4%) carried a defined mutation in the investigated core TERT promoter region (Table 1). The distribution of mutations in the TERT promoter was even in bladder tumors from stages Tis through T2+, ranging from 60% to 71% (Table 1). Low-grade tumors [papillary urothelial tumors of low malignant potential (PUNLMP) + G1] carried mutations in 87 of 118 [73.7%, 95% confidence interval (CI) 65.8–81.7%] and 114 of 180 (63.3%, 95% CI 56.3–70.4%) high-grade tumors (G2 + G3). Thirteen of 29 (44.8%, 95% CI 26.7–62.9%) tumors with undetermined-grade Gx also carried mutations. The distribution of the mutations in different disease categories was similar to those in different tumor stages (Table 1). The DNA sequence amplified for the detection of mutations in TERT promoter also carried a T > C (A > G) single-nucleotide polymorphism (SNP) at position −245 bp (1295349) represented by rs2853669 with a carrier frequency of 47.1% (Table 1). The distribution of somatic mutations in the TERT promoter was not associated with the carrier status of the variant allele of the polymorphism [odds ratio (OR) 1.22, 95% CI 0.79–1.90]. The OR for the distribution of carriers of variant allele according to mutational status was 1.19 (95% CI 0.75–1.88). Of the 327 tumors sequenced for the TERT promoter mutation, data for p53 mutations were available for 174 tumors. Of the 24 tumors with mutations in p53, 17 (71%; 95% CI 53–89%) carried the TERT promoter mutations compared with 94 (63%, 95% CI 55–70%) of 150 that were wild type for p53. The statistical analysis showed an independent association of survival with sex, age at diagnosis, tumor grade, stage, disease categories, presence of node, and metastasis (data not shown). A Cox proportional model, which included age at diagnosis, tumor, node, metastasis status, tumor grade, and treatment, showed a hazard ratio (HR) of 1.34 (95% CI 0.81–2.23) for an overall effect of the mutations on patient survival. The corresponding HR for the survival in patients that were noncarriers of the variant allele of the rs2853669 polymorphism was 2.19 (95% CI 1.02–4.70), and in the carriers, HR was 0.42 (95% CI 0.18–1.01; Table 2). The interaction between the mutations and polymorphism was statistically significant (P = 0.05). The effect of the TERT promoter mutations on survival and its modification by the polymorphism was confirmed in Kaplan–Meier models (Fig. 1). In all subcategories a similar trend of the effect of mutations on patients that did not carry the variant allele was observed (Table 3 and Table S2). 2 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1310522110

The analysis of Tis, Ta, and T1 tumors showed that the TERT promoter mutations in the absence of the variant allele of the polymorphism were associated with statistically significant risk of the disease recurrence (Fig. S3) with an HR of 1.85 (95% CI 1.11–3.08); the association was not statistically significant in the presence of the variant allele (HR 1.14, 95% CI 0.62–1.99). For the patients with TaG1 + TaG2 disease categories, the HR for the risk of the recurrence due to the mutations was 2.53 (95% CI 1.20–5.33) in patients that did not carry the variant allele; the association was not statistically significant in presence of the variant allele (HR 0.80, 95% CI 0.36–1.78; Table 2 and Fig. S4). Luciferase reporter assays were carried out in two urothelial cell carcinoma cell lines, T24 and CLS-439, to determine the effect of the TERT promoter mutations and modulation of the effect by the variant allele of the rs2853669 polymorphism. The results from the assays, carried out in triplicate, showed that the constructs with the variant allele of the polymorphism decreased the promoter activity (t test P = 0.02) compared with the construct with wild-type sequence in both cell lines. Both mutations separately showed twofold to fourfold increased promoter activity compared with wild-type sequences. The introduction of the variant allele of the rs2853669 polymorphism in presence of the mutations resulted in reduction of the promoter activity. The reduction in the activity was statistically significantly in both cell lines (t test; P = 0.03 and 0.05) transfected with the construct with −124C > T mutation/variant allele compared with the construct with only −124C > T mutation. In the construct with −146C > T mutation/variant allele, the reduction in the promoter activity compared with the construct with only −146C > T mutation was moderate and was not statistically significant (Fig. 2). Discussion Bladder carcinoma represents the one of the most frequent cancers worldwide (12, 13). The most common bladder carcinomas are derived from urothelium and are perceived to arise by at least two separate mechanisms. The nonmuscle invasive bladder tumors (Tis, Ta, and T1) account for most of the urothelial carcinomas that are confined to mucosa or submucosa (14). Remaining urothelial carcinomas are invasive that arise either de novo or from high-grade carcinoma in situ (15, 16). Low-grade tumors are characterized by frequent mutations in the HRAS and fibroblast growth factor receptor 3 (FGFR3) genes. A majority of high-grade tumors that show propensity to progression to local and distant metastasis contain defects in p53 and/or retinoblastoma tumor suppressors. The altered pathways in the both variants of bladder cancer have been a subject of research for targeted therapy (17, 18). In this study, we investigated the newly discovered noncoding mutations within the core promoter region of the TERT gene. Two major findings of this study include (i) a high recurrence of the TERT promoter mutations in bladder cancer and (ii) association of the detected mutations with the disease recurrence and patient survival, through an interaction with a common polymorphism. The frequency of the TERT promoter mutations detected was higher than any earlier reported lesions in any gene in bladder cancer (19). A role for the mutations in early bladder carcinogenesis can be deduced from the occurrence in almost two-thirds of bladder tumors across all stages and grades. The specificity of all mutations detected in the TERT promoter in creation of Ets/ TCF binding motif alludes to involvement of a strong selection because telomerase expression is crucial in maintenance of cancer cells in an immortalized state (20). The somatic mutations specifically create CCGGAA/T binding motifs for Ets/TCF, which are distinct from the preexisting GGAA sites in the TERT promoter. The specific mutations detected have been shown to cause twofold to fourfold increased promoter activity in reporter assays (7, 10). Although in melanoma the −146C > T (G > A) was more frequent than the −124C > T (G > A) mutation, in Rachakonda et al.

Table 1. TERT promoter mutations and rs2853669 variant: frequencies in tumors from bladder cancer patients

Patient categories

No.

All Sex Men Women Age [median (Q1 − Q3)= 72.8 y (64.5 − 79.4)] T‡

−146 C > T§

Others{

TT

Carriers (TC + CC)

Percentk

214

65.4

175

38

1

173

154

47.1

221 106

67.6 32.4

146 68

66.1 64.2

120 55

25 13

1 0

119 54

102 52

46.2 49.1

109 189 29

33.3 57.8 8.9

75 124 15

68.8 65.6 51.7

60 101 14

15 22 1

0 1 0

58 102 13

51 87 16

46.8 46.0 55.2

14 104 63 117 29

4.3 31.8 19.3 35.8 8.9

4 83 37 77 13

28.6 79.8 58.7 65.8 44.8

3 68 28 65 11

1 14 9 12 2

0 1 0 0 0

10 51 34 64 14

4 53 29 53 15

28.6 51.0 46.0 45.3 51.7

5 158 56 80 28

1.5 48.3 17.1 24.5 8.6

3 105 40 53 13

60.0 66.5 71.4 66.3 46.4

3 86 32 43 11

0 18 8 10 2

0 1 0 0 0

2 85 30 43 13

3 73 26 37 15

60.0 46.2 46.4 46.3 53.6

30 14 246 37

9.2 4.3 75.2 11.3

20 6 167 21

66.7 42.9 67.9 56.8

16 6 135 18

4 0 31 3

0 0 1 0

18 5 131 19

12 9 115 18

40.0 64.3 46.7 48.6

250 10 32 35

76.5 3.1 9.8 10.7

165 6 24 19

66.0 60.0 75.0 54.3

135 5 18 17

29 1 6 2

1 0 0 0

137 4 15 17

113 6 17 18

45.2 60.0 53.1 51.4

5 101 30 69 80 42

1.5 30.9 9.2 21.1 24.5 12.8

3 80 15 46 53 17

60.0 79.2 50.0 66.7 66.3 40.5

3 65 12 38 43 14

0 14 3 8 10 3

0 1 0 0 0 0

2 50 17 38 43 23

3 51 13 31 37 19

60.0 50.5 43.3 44.9 46.3 45.2

88 196 43

26.9 59.9 13.1

63 126 25

71.6 64.3 58.1

51 102 22

12 23 3

0 1 0

50 102 21

38 94 22

43.2 48.0 51.2

24 150 153

7.3 45.9 46.8

17 94 103

70.8 62.7 67.3

15 76 84

2 18 18

0 0 1

7 83 83

17 67 70

70.8 44.7 45.8

*Percentage of subcategories of patients. † Percentage of the TERT promoter mutations. ‡ C > T or C > A mutation at chr5:1295228 (hg19 coordinate). § C > T mutation at chr5:1295250 (hg19 coordinate). { A > C mutation at chr5:1295161 (hg19 coordinate). k Percentage of carriers of the variant allele for the rs2853669 polymorphism. **PUNLMP: Papillary urothelial neoplasm of low malignant potential. †† Alive or died due to causes than bladder cancer.

bladder tumors the latter mutation was predominant (7). Unlike melanoma, CC > TT tandem mutations were absent in bladder cancer, reflecting etiological differences between the cancer types (21). Our data showed an overall tendency of poor survival in the patients that carried the mutations in tumors. Previously, the Rachakonda et al.

TERT promoter mutations, with prevalence over 80% in gliobastoma, were associated with poor survival (11). A stratification of the data in the present study showed that the effect of the mutations on survival in bladder cancer patients was modified by a common polymorphism rs2853669 located within 120–140 bp of the mutational sites. The mutations did not affect survival in PNAS Early Edition | 3 of 6

GENETICS

Patients

Table 2. Effect of the TERT promoter mutations on patient survival and disease recurrence by Cox regression All Patient categories Survival* Without mutation With mutation Tumor recurrence† Tumor stage (Tis, Ta, T1) Without mutation With mutation Disease category (TaG1 + TaG2) Without mutation With mutation

SNP noncarriers

SNP carriers

N

Dead

HR

(95% CI)

N

Dead

HR

(95% CI)

N

Dead

HR

(95% CI)

93 186

25 63

1.00 1.34

Reference (0.81–2.23)

56 94

13 37

1.00 2.19


37 92

12 26

1.00 0.42


69 138

38 96

1.00 1.51


45 65

22 48

1.00 1.85


24 73

16 48

1.00 1.14


34 89

18 63

1.00 1.59


20 43

9 33

1.00 2.53


14 46

9 30

1.00 0.80


Bold HR values are statistically significant. *HR adjusted for sex, age at diagnosis, TNM status, tumor grade, and treatment. † HR adjusted for sex and age at diagnosis.

the patients that were carriers of the variant allele of the polymorphism. A Cox regression model, which included all factors associated independently with survival including treatment, showed over twofold increased hazard ratio for reduced survival of patients that had the somatic TERT promoter mutations in tumors and were noncarriers of the variant allele. All patients included in the analysis had undergone transurethral resection of bladder tumors; some of them, in addition to the resection, had been treated with combinations of radiotherapy, chemotherapy, and immune therapy. With prevalence of the TERT promoter mutations in two-thirds of bladder cancer, almost one-third of the patients would be vulnerable to an adverse effect on survival according to our data. The nonmuscle invasive lesions, with a propensity to recurrence, are generally not associated with mortality (22). All four patients in the TaG1 disease category, who had died due to the bladder cancer, had the TERT mutations in their tumors and were noncarriers of the variant allele. The data showed that the mutations were also associated with the disease recurrence in patients with Tis, Ta, and T1 tumors. The effect was more pronounced in the absence of the variant allele than in its presence. In the patients with the TaG1 + TaG2 diseases the presence of the mutations and the noncarrier status for the variant allele was associated with a 2.53 increased risk of the tumor recurrence. The current strategy for follow-up treatment is based on risk stratification, dependent on recurrence and progression scores

(14). In clinical trial, bacillus Calmette–Guérin in combination with IFN alpha-2B has been shown to be effective in reducing the recurrence and progression (23). Genetic markers can be of utmost use in stratification of at risk patients. The TERT mutations, particularly in the absence of the variant allele of the polymorphism, can potentially be helpful in identifying patients for intravesical therapies, which are critical for management of nonmuscle invasive bladder cancer (24). Mechanistically, the modifying effect of the polymorphism was in the expected direction. Although the somatic mutations resulted in the creation the Ets/TCF binding sites, the variant allele of the polymorphism disrupted a preexisting noncanonical Ets2 binding site in the proximal region of the TERT promoter, immediately adjacent to an E-box. Interaction of Ets2 with the binding site on TERT promoter has been shown to involve binding of c-Myc to the E-box in breast cancer cell lines (25). Mutation of the Ets2 binding site has been shown to inhibit E-box binding of c-Myc. The variant allele of the SNP has previously been shown to be associated with low telomerase and telomere activity in non-small cell lung carcinoma (26). The reported association of the polymorphisms with risk of inherited susceptibility to breast cancer has been inconsistent; however, the tissue and cellular context could be important for functionality as observed in reporter assays (27–30). Based on the information from the 1000 Genomes database (www.1000genomes.org), the rs2853669 polymorphism is in linkage

Fig. 1. Kaplan–Meier analysis of differences in survival (A) in all patients with and without the TERT promoter mutations; (B) in patients that were noncarriers of the variant allele rs2853669, with and without the mutations; and (C) in patients that were carriers of the variant allele rs2853669, with and without the mutations. The numbers of patients at risk in each category are shown along the curves at years 5, 10, and 15.

4 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1310522110

Rachakonda et al.

Table 3. Cox regression analysis for the effect of the TERT promoter mutations and rs2853669 polymorphism on cause-specific survival in subcategories of bladder cancer patients All Patients Men Without mutation With mutation Women Without mutation With mutation Age at diagnosis ≥70 y Without mutation With mutation Age at diagnosis C (T > G; 1295161) as a somatic mutation in a bladder tumor, which we earlier reported as a highpenetrant causal mutation in a large melanoma pedigree, reinforces the status of the nucleotide changes in the TERT promoter as bonafide driver mutations. p53 has been shown to repress TERT transcription in a Sp1-dependent manner. One of

the tumors with a single nucleotide deletion together with the −146C > T mutation represented a unique case where the targeted mutations resulted in an ablation of as Sp1 repressor element and creation of a new site for Ets/TCF (4, 33). The resulting increased expression of telomerase as a consequence of the TERT promoter mutations can be a target for therapy. Several strategies of therapeutic telomerase inhibition, including small molecular inhibitors, immunotherapy, gene therapy, and telomere- and telomerase-associated proteins, in different cancers have entered clinical trials (34). The telomerase protein is also associated with other biological activities that include enhanced cell proliferation, inhibition of apoptosis, and regulation of DNA damage response and cellular proliferative life span that could be affected by the promoter mutations (35). At a wider level, the promoter mutations represent a paradigm shift in the trail of missing mutations that could not be accounted through the search confined to coding sequences (36). In the context of bladder cancer, we show that the mutations affect the survival in a manner that is modified by a common genetic polymorphism located within the same locus. The mutations and the polymorphism in conjunction can potentially be used as clinical biomarkers to predict the patient survival and the disease recurrence.

Fig. 2. Relative luciferase activity of different TERT promoter constructs. Luciferase reporter constructs containing wild-type (WT) TERT promoter sequence, sequence with variant allele for the rs2853669 polymorphism (WT+rs), with −124T mutant (−124T), with −124T mutant and the variant allele (−124T+rs), with −146T mutant (−146T), and with −146 mutant and the variant allele (−146T+rs) were transfected into (Left) T24 and (Right) CLS-439 urothelial carcinoma cell lines. Constructs with either of the two mutations (−124T and −146T) resulted in twofold to fourfold increased activity compared with the construct with WT. Constructs with −124T+rs or −146T+rs showed reduced activity in comparison with constructs with corresponding mutations only.

Rachakonda et al.

PNAS Early Edition | 5 of 6

GENETICS

Bold HR values are statistically significant. *HR adjusted for sex, age at diagnosis, TNM status, tumor grade, and treatment.

Materials and Methods Patients. A total of 538 patients were registered in the Stockholm region with newly diagnosed bladder neoplasms (urothelial cell carcinomas) in 1995–1996 with 15 y clinical follow-up. All hospitals and urology units in the region participated in the study. Treatment and follow-up were performed according to a standard-of-care program. Tumors were removed from patients, before any treatment, with transurethral resection; snap frozen in liquid nitrogen; and cut in approximately 5-μm-thick sections. The first and last sections were stained and examined for tumor content. Only biopsies with more than 70% tumor were included in the study. Tumor stage was assessed according to modified tumor, node, metastasis (TNM) system suggested by Hall and Prout (37). The World Health Organization 1999 malignancy grading system was used for tumor grade (38). Informed consent was obtained from all patients included in the study. The study was approved by the Regional Ethics Committee of Stockholm at Karolinska Institutet. In this study, 327 patients were included, from whom tumor DNA was either available or amplified successfully in PCR. The treatment data were available for 284 patients. One hundred forty-three patients were treated with transurethral resection of bladder tumor only, whereas other patients received different treatment combinations (Table S3). Mutation Detection. DNA was extracted from tumors and corresponding blood tissues using standard procedures. The core TERT promoter region 343 bp (from the position −278 to +65 from ATG start site) was screened for mutations using PCR and Sanger sequencing (Table S4) as described (7). More details are provided in SI Materials and Methods. Mutations in the p53

1. Dwyer J, Li H, Xu D, Liu JP (2007) Transcriptional regulation of telomerase activity: Roles of the the Ets transcription factor family. Ann N Y Acad Sci 1114:36–47. 2. Daniel M, Peek GW, Tollefsbol TO (2012) Regulation of the human catalytic subunit of telomerase (hTERT). Gene 498(2):135–146. 3. Hanahan D, Weinberg RA (2011) Hallmarks of cancer: The next generation. Cell 144(5):646–674. 4. Kyo S, Takakura M, Fujiwara T, Inoue M (2008) Understanding and exploiting hTERT promoter regulation for diagnosis and treatment of human cancers. Cancer Sci 99(8): 1528–1538. 5. Stein GS, Montecino M, van Wijnen AJ, Stein JL, Lian JB (2000) Nuclear structure-gene expression interrelationships: Implications for aberrant gene expression in cancer. Cancer Res 60(8):2067–2076. 6. Renaud S, et al. (2007) Dual role of DNA methylation inside and outside of CTCFbinding regions in the transcriptional regulation of the telomerase hTERT gene. Nucleic Acids Res 35(4):1245–1256. 7. Horn S, et al. (2013) TERT promoter mutations in familial and sporadic melanoma. Science 339(6122):959–961. 8. Wei GH, et al. (2010) Genome-wide analysis of ETS-family DNA-binding in vitro and in vivo. EMBO J 29(13):2147–2160. 9. Patton EE, Harrington L (2013) Cancer: Trouble upstream. Nature 495(7441):320–321. 10. Huang FW, et al. (2013) Highly recurrent TERT promoter mutations in human melanoma. Science 339(6122):957–959. 11. Killela PJ, et al. (2013) TERT promoter mutations occur frequently in gliomas and a subset of tumors derived from cells with low rates of self-renewal. Proc Natl Acad Sci USA 110(15):6021–6026. 12. Burger M, et al. (2013) Epidemiology and risk factors of urothelial bladder cancer. Eur Urol 63(2):234–241. 13. Cohen SM, Shirai T, Steineck G (2000) Epidemiology and etiology of premalignant and malignant urothelial changes. Scand J Urol Nephrol Suppl (205):105–115. 14. Anastasiadis A, et al. (2012) Follow-up procedures for non-muscle-invasive bladder cancer: An update. Expert Rev Anticancer Ther 12(9):1229–1241. 15. Wu XR (2005) Urothelial tumorigenesis: A tale of divergent pathways. Nat Rev Cancer 5(9):713–725. 16. Cheng L, et al. (2011) Bladder cancer: Translating molecular genetic insights into clinical practice. Hum Pathol 42(4):455–481. 17. Gschwind A, Fischer OM, Ullrich A (2004) The discovery of receptor tyrosine kinases: Targets for cancer therapy. Nat Rev Cancer 4(5):361–370. 18. Kuball J, et al. (2002) Successful adenovirus-mediated wild-type p53 gene transfer in patients with bladder cancer by intravesical vector instillation. J Clin Oncol 20(4): 957–965. 19. Dulak AM, et al. (2013) Exome and whole-genome sequencing of esophageal adenocarcinoma identifies recurrent driver events and mutational complexity. Nat Genet 45(5):478–486. 20. Hahn WC, et al. (1999) Creation of human tumour cells with defined genetic elements. Nature 400(6743):464–468. 21. Thilly WG (2003) Have environmental mutagens caused oncomutations in people? Nat Genet 34(3):255–259. 22. Prasad SM, Decastro GJ, Steinberg GD; Medscape (2011) Urothelial carcinoma of the bladder: Definition, treatment and future efforts. Nat Rev Urol 8(11):631–642.

6 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1310522110

gene (exon 5–8) in a subset of the bladder tumors included in this study were determined and described previously (39, 40). Statistical Analysis. The associations between the age, sex, p53 mutational status, stage, grade, node, metastasis, TERT promoter mutations, and causespecific survival were determined separately using the Kaplan–Meier method. The cumulative curves for the survival outcome in patients with and without the TERT promoter mutations, before and after stratification into carriers and noncarriers of the variant allele of the rs2853669 polymorphism, were drawn. The differences between the curves were analyzed with log rank test. A Cox proportional hazard regression model was used to determine the association between the TERT promoter mutations, before and after stratification for carriers and noncarriers of the variant alleles of the rs2853669 polymorphism, and cause-specific patient survival. The model was adjusted for sex, age at diagnosis, tumor stage, grade, node, metastasis, and treatment. Five patients with Tis were excluded from survival analysis. A Cox model was also used to determine the association between the TERT promoter mutations and the disease recurrence in patients with noninvasive tumors, both before and after stratification for the variant allele of the polymorphism. Luciferase Reporter Assays. Luciferase reporter assays were carried out by transfecting two uroithelial carcinoma cell lines, T24 and CLS-439, using different reporter constructs in triplicate. More details about the cell lines, cloning, and reporter assays are provided in SI Materials and Methods. ACKNOWLEDGMENTS. We thank Ms. Antje Sucker for TERT promoter constructs.

23. Joudi FN, Smith BJ, O’Donnell MA; National BCG-Interferon Phase 2 Investigator Group (2006) Final results from a national multicenter phase II trial of combination bacillus Calmette-Guérin plus interferon alpha-2B for reducing recurrence of superficial bladder cancer. Urol Oncol 24(4):344–348. 24. Logan C, Brown M, Hayne D (2012) Intravesical therapies for bladder cancer— Indications and limitations. BJU Int 110(Suppl 4):12–21. 25. Xu D, Dwyer J, Li H, Duan W, Liu JP (2008) Ets2 maintains hTERT gene expression and breast cancer cell proliferation by interacting with c-Myc. J Biol Chem 283(35): 23567–23580. 26. Hsu CP, Hsu NY, Lee LW, Ko JL (2006) Ets2 binding site single nucleotide polymorphism at the hTERT gene promoter—Effect on telomerase expression and telomere length maintenance in non-small cell lung cancer. Eur J Cancer 42(10): 1466–1474. 27. Savage SA, et al. (2007) Genetic variation in five genes important in telomere biology and risk for breast cancer. Br J Cancer 97(6):832–836. 28. Varadi V, et al. (2009) A functional promoter polymorphism in the TERT gene does not affect inherited susceptibility to breast cancer. Cancer Genet Cytogenet 190(2): 71–74. 29. Beesley J, et al.; kConFab Investigators; Australian Ovarian Cancer Study Group; ABCTB Investigators; Ovarian Cancer Association Consortium (2011) Functional polymorphisms in the TERT promoter are associated with risk of serous epithelial ovarian and breast cancers. PLoS ONE 6(9):e24987. 30. Kote-Jarai Z, et al.; COGS-CRUK GWAS-ELLIPSE (Part of GAME-ON) Initiative; UK Genetic Prostate Cancer Study Collaborators/British Association of Urological Surgeons’ Section of Oncology; UK ProtecT Study Collaborators; PRACTICAL Consortium (2013) Fine-mapping identifies multiple prostate cancer risk loci at 5p15, one of which associates with TERT expression. Hum Mol Genet 22(12):2520–2528. 31. Abecasis GR, et al.; 1000 Genomes Project Consortium (2012) An integrated map of genetic variation from 1,092 human genomes. Nature 491(7422):56–65. 32. Bojesen SE, et al. (2013) Multiple independent variants at the TERT locus are associated with telomere length and risks of breast and ovarian cancer. Nat Genet 45(4): 371–384. 33. Xu D, et al. (2000) Downregulation of telomerase reverse transcriptase mRNA expression by wild type p53 in human tumor cells. Oncogene 19(45):5123–5133. 34. Ruden M, Puri N (2013) Novel anticancer therapeutics targeting telomerase. Cancer Treat Rev 39(5):444–456. 35. Mukherjee S, Firpo EJ, Wang Y, Roberts JM (2011) Separation of telomerase functions by reverse genetics. Proc Natl Acad Sci USA 108(50):E1363–E1371. 36. Vogelstein B, et al. (2013) Cancer genome landscapes. Science 339(6127):1546–1558. 37. Hall RR, Prout GR (1990) Staging of bladder cancer: Is the tumor, node, metastasis system adequate? Semin Oncol 17(5):517–523. 38. Reuter VE, Epstein JI, Amin MB, Mostofi FK (1999) The “WHO/ISUP Consensus Classification of Urothelial (Transitional Cell) Neoplasms”: Continued discussion. Hum Pathol 30(7):879–880. 39. Ryk C, et al. (2005) Influence of GSTM1, GSTT1, GSTP1 and NAT2 genotypes on the p53 mutational spectrum in bladder tumours. Int J Cancer 113(5):761–768. 40. Berggren P, et al. (2001) p53 mutations in urinary bladder cancer. Br J Cancer 84(11): 1505–1511.

Rachakonda et al.