TERT Promoter Mutations Frequency Across Race, Sex, and Cancer Type

Abstract

Background

Telomerase reverse transcriptase (TERT) gene promoter mutations have been explored, as biomarkers of improved survival for patients with cancer receiving immune checkpoint inhibitors. We sought to investigate their prevalence by race and sex across different cancer types to inform patient selection in clinical trials.

Results

In this observational study, 31 925 patients with cancer underwent next-generation sequencing of their tumors with 88% (27 970) patients self-reported being Whites, 7.1% (2273) Asians, and 5.3% (1682) Blacks. Examining the distribution of TERT promoter mutations by race, White patients with melanoma harbored more TERT promoter mutations than Asian and Black patients (OR = 25.83; 95%CI, 6.84-217.42; P < .001). In contrast, Asian patients with head and neck cancer (HNC) harbored more TERT promoter mutations compared to White patients (OR = 2.47; 95%CI, 1.39-4.37; P = .004). In addition, the distribution of TERT promoter mutations differed by sex. Males were enriched for TERT gene promoter mutations compared to females with melanoma (OR = 1.82; 95%CI, 1.53-2.16; P < .001), cancer of unknown primary (OR = 1.96; 95%CI, 1.43-2.69; P < .001), hepatobiliary (OR = 3.89; 95%CI, 2.65-5.69; P < .001), and thyroid cancers (OR = 1.42; 95%CI, 1.10-1.84; P = .0087), while females were more enriched for TERT promoter mutations compared to males for HNC (OR = 0.56; 95%CI, 0.39-0.81; P = .0021).

Conclusions

The prevalence of TERT gene promoter mutations varies among patients with cancer based on race and sex. These findings inform our understanding of cancer biology and can assist in the design of future clinical trials that leverage drugs targeting TERT promoter dependencies.

Introduction

Telomerase is a reverse transcriptase that synthesizes the 3ʹ end of linear chromosomes.^1,2 In human somatic cells, telomeres shorten with each cellular replication, as DNA polymerase cannot fully replicate the 3ʹ end of DNA.^3,4 Telomere shortening is a cardinal feature of the aging process, where cells undergo senescence, apoptosis, or genomic instability.⁵ Telomerase counteracts this process by adding telomeres and is inhibited in most normal human cells across different tissue types. Exceptions include continually regenerating cells (eg, stratum basalis of the skin, testes, or ovaries) and cells seeking to maintain cell proliferation and avoid senescence (ie, cancer cells).^6,7

The telomerase reverse transcriptase (TERT) gene encodes the enzymatic subunit of the telomerase complex.⁸ In cancer cells, activating mutations in the TERT promoter region can increase TERT protein production and the subsequent activity of telomerase, resulting in increased telomere length.⁹ Many studies have investigated the regulation of TERT in healthy tissues and its aberrant expression in malignant cells.^10-13 In addition, by promoting the epithelial-mesenchymal transition,¹⁴TERT promoter mutations can increase PD-L1 expression¹⁵ and thus increase response to ICI therapy. While genetic risk assessments in the clinic do not routinely test for TERT promoter mutations, currently, the notion is gaining traction as a strategy for helping determine prognosis.¹⁶ In the era of immune checkpoint inhibitors (ICIs), De Kouchkovsky et al¹⁷ recently suggested that the presence of TERT promoter mutations in urothelial cancer was a predictor of improved overall survival in a cohort of 119 patients treated with ICIs. In addition, a recent pan-cancer analysis of 10 336 patients sequenced using MSK-IMPACT showed variability in the prevalence of TERT mutations across cancer types and found an association between the presence of TERT promoter mutations and higher tumor mutational burden (TMB) and neoantigen load. The authors also noted better overall survival outcomes with anti-CTLA4 therapy, among patients with melanoma harboring TERT promoter mutations.¹⁸

Given the importance of TERT promoter mutations in cancer development and their potential use as a biomarker for clinical outcomes and response to treatment with ICIs, we investigated the distribution of TERT promoter mutations by sex and race across different solid tumors. Their prevalence across specific groups of patients can potentially guide future patient selection in clinical trials and therapeutic strategies.

Methods

Study Design

Patients diagnosed with 1 of the 10 solid tumors most commonly harboring TERT gene promoter mutations (thyroid cancer, soft-tissue sarcoma, non-small cell lung cancer (NSCLC), melanoma, hepatobiliary cancer (HPBC), head and neck cancer (HNC), glioma, colorectal cancer, cancer of unknown primary, and bladder cancer) were identified from the American Association for Cancer Research (AACR) Project Genomics Evidence Neoplasia Information Exchange (GENIE) 11.0-registry.¹⁹ As the bait-set coverage of the TERT gene was available from Dana-Farber Cancer Institute (DFCI) and Memorial Sloan Kettering Cancer Center (MSKCC), we selected patients from these 2 cancer centers for this analysis. All patients underwent next-generation sequencing of tumors with CLIA-/ISO-certified panels obtained from the respective institution from which they were recruited. Mutational analyses in this study were focused on TERT promoter mutations. Patients whose tumors were sequenced using DFCI OncoPanel version 1 were excluded, because it does not cover the entire TERT gene sequence. Methodology for tissue collection, DNA extraction and sequencing using the DFCI-ONCOPANEL, and MSK-IMPACT for DFCI and MSKCC patients, respectively, were previously described in detail.^20-23 Additional details on methodology are provided in Supplementary material.

Statistical Analysis

The prevalence of TERT promoter mutations was summarized as “N” and percentage by sex and race across different solid tumors. The strength of associations was quantified as the odds ratio (OR) with 95% confidence intervals (95%); 2-sided Fisher’s exact tests were applied. FDR-adjusted Q-values were also calculated to correct P-values using the Benjamini-Hochberg method. P-values < .05 were considered significant as long as they maintained a Q-value < .05 on multiple test corrections.

Results

A total of 31 925 patients from DFCI and MSKCC were included (Table 1; Supplementary Table S1). In our cohort, 88% (27 970) patients self-reported being Whites, 7.1% (2273) Asians, and 5.3% (1682) Blacks. TERT promoter mutations were identified in 56% (1346/2411) bladder cancers, 43% (966/2243) melanomas, 39% (1403/3614) gliomas, 41% (395/959) thyroid cancers, and 23% (183/782) HNC (Fig. 1; Supplementary Table S2).

Examining the differences in the distribution of TERT promoter mutations by race, it was notable that there were differential distributions among patients with the same cancer type but belonging to different racial groups (Supplementary Tables S3 and S4). For example, White patients with melanoma (44%, n = 964/2176) harbored more TERT promoter mutations compared to Asian and Black patients (3%, n = 2/67) (OR = 25.83; 95%CI, 6.84-217.42; Q < .001) (Supplementary Table S4). Similarly, TERT promoter mutations were differentially enriched among patients by sex (Supplementary Table S5). TERT promoter mutations were significantly more frequent among male patients as compared to their female counterparts with melanoma (49%, n = 638/1298 vs. 35%, n = 328/945, respectively) (OR = 1.82; 95%CI, 1.53-2.16; Q < .001), HPBC (17%, n = 137/827 vs. 4.9%, n = 36/741, respectively) (OR = 3.89; 95%CI, 2.65-5.69; Q < .001), cancer of unknown primary (11%, n = 116/1065 vs. 5.9%, n = 64/1081, respectively) (OR = 1.96; 95%CI, 1.43-2.69; Q < .001), and thyroid cancer (46%, n = 213/468 vs. 37%, n = 182/491, respectively) (OR = 1.42; 95%CI, 1.10-1.84; Q = .017). In contrast, male patients with HNC (21%, n = 118/575) harbored significantly fewer TERT promoter mutations than female patients (31%, n = 65/207) (OR = 0.56; 95%CI, 0.39-0.81; Q = .005) (Fig. 2). When examining patients by ethnicity, there were no significant differences in the distribution of TERT promoter mutations between Hispanic and non-Hispanic patients (Supplementary Table S6).

Furthermore, TERT promoter mutations differed between males and females within the same race or between different racial groups in patients with the same sex. For example, TERT promoter mutations occurred more often in White males with HPBC (16%, n = 112/692) compared to their female counterparts (4.9%, n = 31/635) (OR = 3.76; 95%CI, 2.49-5.69; Q < .001), and Black males (28%, n = 13/46) harbored more TERT promoter mutations than Black females (4.1%, n = 2/49) (OR = 9.26; 95%CI, 1.96-43.9; Q = 0.008) (Fig. 3; Supplementary Table S7). Whereas in patients with HNC, Asian males (37%, n = 22/60) were significantly more enriched for TERT promoter mutations relative to White males (19%, n = 93/489) (OR = 2.47; 95%CI, 1.39-4.37; P = .004) (Fig. 3; Supplementary Table S7). No significant differences in TERT promoter mutations distribution were determined across racial and sex subgroups among patients with glioma, soft-tissue sarcoma, bladder, colorectal, NSCLC, and thyroid cancers (Fig. 3; Supplementary Table S7). Further exploration of HNC tumors revealed that oral cavity squamous cell carcinoma (OSCC) tumors (44%, n = 85/192) were associated with higher TERT promoter mutations vs. other HNC subtypes (17%, n = 98/590) (OR = 3.98; 95%CI, 2.74-5.79; Q = .03). Asian patients with OSCC (70%, n = 14/20) harbored more mutations vs. non-Asian patients with OSCC (41%, n = 71/172) (OR = 3.3; 95%CI, 1.2-9.0; P = .019) (Supplementary Table S8).

Finally, the distribution of TERT promoter mutations was compared between primary and metastatic tumor samples. Mutations were significantly more frequent among samples from primary tumors as compared with samples from metastatic tumors in patients with bladder cancer (58.4%, n = 1076/1843 vs. 38.3%, n = 270/559, respectively) (OR = 1.5; 95%CI, 1.2-1.8; Q < .001), cancer of unknown primary (12.6%, n = 33/261 vs. 8.2%, n = 133/1613, respectively) (OR = 1.6; 95%CI, 1.1-2.4; Q = .0248), colorectal cancer (4.5%, n = 191/4219 vs. 2.5%, n = 51/2053, respectively) (OR = 1.9, 95%CI, 1.4-2.6; Q < .001), glioma (42.3%, n = 1287/3039 vs. 25.3%, n = 102/403, respectively) (OR = 2.2; 95%CI, 1.7-2.7; Q < .001), HNC (27.4%, n = 116/423 vs. 19%, n = 67/352, respectively) (OR = 1.6; 95%CI, 1.1-2.3; Q = .0091), HPBC (13.3%, n = 148/966 vs. 5.8%, n = 25/433, respectively) (OR = 2.5; 95%CI, 1.6-3.9; Q < 0.001). However, samples from primary tumors were significantly less enriched with TERT promoter mutations than those from metastatic tumors in patients with melanoma (31%, n = 223/720 vs. 49.6%, n = 741/1495, respectively) (OR = 0.5; 95%CI, 0.4-0.6; Q < .001), soft-tissue sarcoma (6.3%, n = 99/1578 vs. 8.9%, n = 75/838, respectively) (OR = 0.7, 95%CI, 0.5-0.9; Q = .03), and thyroid cancer (35.2%, n = 177/503 vs. 48%, n = 217/452, respectively) (OR = 0.6; 95%CI, 0.5-0.8; Q < .001) (Supplementary Fig. S1; Supplementary Table S9,). The distribution of TERT promoter mutations were also evaluated across race and sex groups in primary (Supplementary Table S10) and metastatic (Supplementary Table S11) samples, respectively.

Discussion

In this study, we report for the first time, differences in the distribution of TERT promoter mutations by race and sex across different cancer types. Our findings reveal that TERT promoter mutations are frequent events in many cancer types, including bladder cancer, melanoma, thyroid cancer, glioma, and HNC. TERT promoter mutations are found to be more prevalent among male patients than their female counterparts with most cancer types, except for HNC where the opposite held true. Additional noteworthy results include a more significant enrichment among patients of White race with melanoma and patients of Asian race with HNC, particularly OSCC.

The prevalence of TERT promoter mutations in melanoma varies between 29% and 71%.^24,25 In our cohort, 43% of 2243 melanoma cases harbored TERT promoter mutations. A higher frequency of TERT mutations among male patients in melanoma compared with their female counterparts has been described.²⁶ In addition, our findings reveal ethnic differences in this disease, with patients of White race carrying significantly more TERT promoter mutations compared to other groups. Variations in skin color and the protective effects of darker skin against ultraviolet radiation may counterpart some of this difference.²⁷ Griewank et al²⁸ found a considerably higher frequency of TERT promoter mutations in cutaneous melanomas of UV-exposure-prone areas than in tumors arising in areas with minimal or absent sun exposure.

Other factors may be implicated in these differences based on race. For example, TERT promoter mutations have been associated with increased patient age at diagnosis, as well as increased Breslow thickness, tumor ulceration, and tumor growth rate in patients with melanoma.^28-31 In keeping with this, more adverse outcomes and worse survival have been consistently reported among patients with melanoma who harbor TERT promoter mutations compared to those who do not.^28,32,33 However, it is notable that Black patients have a higher prevalence of advanced cutaneous melanoma³⁴ with worse pathologic features (ie, tumor ulceration and satellite nodules) and regional and distant metastases.^35-37 Rouhani et al³⁸ demonstrated that Black patients presented with a significantly later stage at their time of diagnosis with melanoma compared with White American patients. As such, Black patients might also have worse outcomes due to underlying health disparities and access to care that may have led to delays in diagnosis and treatment compared to groups of patients who have more access to screening, such as skin examinations.³⁹ This may explain why White patients continue to demonstrate better overall survival than non-White groups across different melanoma subtypes, irrespective of TERT promoter mutations.^34,36,40

Regarding HNC, Yu et al⁴¹ found a mutation rate of 81% in 74 patients with OSCC. TERT promoter mutations were also found to be associated with an increased risk of locoregional failure (subdistribution HR = 2.82; 95%CI, 1.47-5.42; P = .0019) independent of oral cavity primary site and TP53 mutation status. Our study included a larger sample size and found a mutation rate of 44.27% of 192 OSCC with a frequency of 70% in OSCC in Asian patients (n = 20).

While certain cancer types (ie, HNC and HPBC) exhibited a higher rate of TERT promoter mutations in primary tumor samples compared to metastatic samples, mutations were more prevalent in metastatic samples of melanoma and thyroid cancer compared to primary tumors. Prior work by Morris et al⁴² showed that metastatic HNC tumors demonstrated a higher prevalence of TERT promoter mutations compared to primary tumors. Whereas Hugdahl et al⁴³ found no significant difference in mutation distribution between primary and metastatic melanoma tumors. In addition, Yang et al⁴⁴ showed that TERT mutant thyroid cancers were linked to a higher occurrence of distant metastasis at the time of diagnosis. Given that primary and metastatic samples may have different genomic alterations, it is imperative that future studies highlight the development of TERT promoter mutations throughout the course of tumor development and its progression to further understand the relationship of these mutations with tumor aggressiveness and spread.

Overall, our results add to the growing body of the literature that serves to enhance our understanding of the variation in TERT promoter mutations across cancer types, sex, and race. Our study has several limitations, as it included patients from tertiary academic centers who may be enriched for unique features based on referral or geographic patterns. In addition, the analyses performed relied on subjective self-reported race, rather than objective parameters (ie, ancestry). However, ancestry-based approaches have similarly found differences in the rate of TERT promoter mutations across ancestral groups. Arora et al⁴⁵ recently demonstrated that patients with glioblastoma multiforme and hepatocellular carcinoma with east Asian ancestry had lower frequency of TERT promoter mutations compared to patients with European ancestry. Moreover, data on TMB were not available, and we were unable to adjust for it when examining the incidence of TERT promoter mutations across subgroups. Patients with high TMB are more likely to display genomic alterations,⁴⁶ which may have affected their distribution across different groups. It is also noteworthy to highlight that the reporting of TERT promoter mutations may be significantly influenced by intratumoral heterogeneity especially when examining differences between localized tumor and metastatic specimens. Finally, heterogeneity in tumor subtypes which may have not been fully reported (ie, melanoma) may have introduced bias into our subgroups and subsequent analyses.

In conclusion, our study demonstrates distinct differences in their distribution by sex and race across several cancer types. It will be essential to account for these variations when launching future trials to assess the impact of these mutations on clinical outcomes and their use as a biomarker.

Funding

None.

Conflict of Interest

T.E.Z. received honoraria from art tempi for work unrelated to this work. W.X. serves as consultant for Convergent Therapeutics, Inc. T.K.C. reports institutional and/or personal, paid and/or unpaid support for research, advisory boards, consultancy, and honoraria from Alkermes, AstraZeneca, Aravive, Aveo, Bayer, Bristol Myers-Squibb, Calithera, Circle Pharma, Eisai, EMD Serono, Exelixis, GlaxoSmithKline, Gilead, IQVA, Infinity, Ipsen, Jansen, Kanaph, Lilly, Merck, Nikang, Nuscan, Novartis, Pfizer, Roche, Sanofi/Aventis, Scholar Rock, Surface Oncology, Takeda, Tempest, Up-To-Date, CME events (Peerview, OncLive, MJH, and others), outside the submitted work. Institutional patents filed on molecular alterations and immunotherapy response/toxicity, and ctDNA. Equity: Tempest, Pionyr, Osel, Precede Bio, CureResponse. Committees: NCCN, GU Steering Committee, ASCO/ESMO, ACCRU, KidneyCan. Medical writing and editorial assistance support may have been funded by Communications companies in part. No speaker’s bureau. Mentored several non-US citizens on research projects with potential funding (in part) from non-US sources/Foreign Components. The institution (Dana-Farber Cancer Institute) may have received additional independent funding of drug companies or/and royalties potentially involved in research around the subject matter. T.K.C. is supported in part by the Dana-Farber/Harvard Cancer Center Kidney SPORE (2P50CA101942-16) and Program 5P30CA006516-56, the Kohlberg Chair at Harvard Medical School and the Trust Family, Michael Brigham, Pan Mass Challenge, Hinda and Arthur Marcus Fund, and Loker Pinard Funds for Kidney Cancer Research at DFCI. The other authors indicated no financial relationships.

Author Contributions

Conception/design: all authors. Provision of study material or patients: T.E.Z., M.M., R.N. Collection and/or assembly of data: T.E.Z., M.M., R.N. Data analysis and interpretation: all authors. Manuscript writing and final approval of manuscript: all authors.

Data Availability

The data underlying this article are available in the article and in Supplementary material.

References

Author notes