Disease-Specific and Histological Outcomes of Preoperative TERT Promoter Mutation Detected in Isolation or in Combination with Other Mutations in Thyroid Neoplasms

Abstract

Background:

TERT promoter mutations in thyroid cancer are associated with aggressive disease, including recurrence, distant metastasis, and disease-related mortality. We aim to assess histological and disease-related outcomes when TERT mutation is detected alone or with other alterations during preoperative testing.

Methods:

A retrospective, single-institution study was performed, including all adult patients undergoing initial diagnostic thyroid nodule evaluation with TERT promoter mutation (C228T/C250T) detected in preoperative thyroid fine needle aspiration samples using TSv3 testing.

Results:

Of 70 thyroid nodules, 18 (26%) were isolated (iTERT), and 52 (74%) were associated with ≥1 concurrently detected mutation (TERT+). The most common additional abnormalities were BRAF^V600E (23, 44%), RAS (19, 37%), and copy number alterations (CNA; 15, 29%). Patients with iTERT were older than those with TERT+ nodules (p = 0.007). While nodule size was similar between the two groups (mean size 3.2 cm, p = 0.18), Bethesda III/IV cytology was more likely with iTERT (94% vs. Bethesda V/VI 56%, p = 0.007). Histology was available for 9 (50%) iTERT and 51 (98%) TERT+ nodules and malignancy was higher with preoperative detection of TERT+ compared with iTERT (96% vs. 67%, p = 0.02). Poorly differentiated or anaplastic cancers were diagnosed in 33% of the malignancies at an equivalent rate in both cohorts. At median follow-up of 13.1 months (interquartile range 26.2, 7.2–33.4), distant metastasis occurred in 32.7% of patients including 17% (1/6) with iTERT versus 33% of patients with TERT+ (p = 0.65). None of the iTERT patients had locoregional recurrence as compared with 25% of TERT+ patients (p = 0.31).

Conclusions:

When preoperatively detected, TERT promoter mutations are more often seen in association with additional driver mutations or other genetic alterations such as CNA, but when present, carries a very high risk of malignancy (93%). Up to 1/3 are poorly differentiated or anaplastic thyroid cancer, and this likelihood is equivalent when TERT is present in isolation or in combination with other mutation. Thus, surgery should be strongly considered in iTERT nodules, and total thyroidectomy should be favored when TERT+ is identified.

Introduction

Thyroid cytology can accurately differentiate thyroid nodules as benign or malignant; however, a substantial percentage of nodules (10–38%) have indeterminate cytology (Bethesda III–VI).¹ Molecular testing may be used to further risk stratify, including avoiding the need for surgery in nodules that are likely to be benign.^2,3 The specific molecular alteration has also been demonstrated to potentially provide prognostic information.^4,5

In 2013, the TERT promoter mutation was first associated with thyroid cancer and has been associated with cancer aggressiveness, including recurrence, distant metastasis, and disease-related mortality.^6,7 TERT mutations are also seen with progression of dedifferentiation, with a prevalence of up to 43.2% and 40.1% in poorly differentiated thyroid cancers (PDTC) and anaplastic thyroid cancer (ATC), respectively.⁸ The overall incidence of TERT promoter mutation in thyroid cancer has been reported to be 4–6%.^9,10 The mutation is a point mutation (either C250T or C228T) and is associated with overproduction of telomerase, an enzyme that is not typically the initial driver toward tumorigenesis but may be a mechanism for cells to evade apoptosis.¹¹

Given the aggressiveness of cancers associated with TERT promoter mutations, surgical intervention, if feasible, is often recommended. However, most studies to date have selectively assessed for TERT on histology, and no study has specifically assessed outcomes when TERT is detected preoperatively. A recent meta-analysis included 77 TERT mutated thyroid nodules from 26 different manuscripts (2014–2022) to identify an 87% cancer risk when TERT is isolated compared with 85% when TERT is identified with RAS and 100% when TERT and BRAF^V600E are both present.¹² The aim of this study is to assess the histological and disease-related outcomes when TERT promoter mutation is detected during preoperative testing under a uniform testing algorithm and in the era of noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP).

Methods

A retrospective study was performed, including all ThyroSeq version 3 (TSv3) results that were positive for TERT promoter mutation (either C250T or C228T) during testing of fine needle aspiration (FNA) specimens of a thyroid nodule undergoing evaluation for consideration of index thyroidectomy from March 2017 to January 2023. Molecular testing was performed on the residual specimen collected during FNA sampling using next-generation sequencing as described previously.⁵ Indications for preoperative molecular testing of FNA biopsy results followed guideline recommendations and included Bethesda III/IV/V thyroid nodule cytology when results would change management, and Bethesda VI nodules when further risk stratification is needed for decision-making, such as when neoadjuvant therapy would be clinically appropriate. Since results were obtained when needed for clinical management, these results were available to clinicians for subsequent decision-making.^3,13 FNA specimens of radiographical abnormalities identified during surveillance for an already treated primary thyroid cancer were excluded. Patient demographics, radiographical findings, pathological results, clinical outcomes, and treatment data were obtained via chart review.

A post hoc analysis was conducted to further evaluate nodules initially classified as nonmalignant despite detection of TERT promoter mutations. Archived histopathology slides were retrieved for review by an experienced endocrine pathologist. For cases with available material, histological features were reassessed for evidence of malignancy. Additionally, repeat TSv3 testing was used to confirm the presence of TERT promoter mutations.

Statistical tests were performed using GraphPad Prism version 10.0.0 for Windows (GraphPad Software, Boston, Massachusetts, USA, www.graphpad.com). Chi-square analysis or Fisher’s exact test was used for categorical variables, and continuous variables were analyzed with Kruskal–Wallis tests. The Kaplan–Meier method was performed to evaluate recurrence-free survival using SPSS (IBM SPSS, version 29.0.1.0).

This study involving human subjects was reviewed and approved by the University of Pittsburgh Medical Center (UPMC) Institutional Review Board (QI-IRB #4438). Written informed consent from the participants was not required per UPMC IRB. The study followed appropriate processes laid out by the IRB.

Results

Initially, 76 patients were identified with preoperative TERT identified in FNA specimens during the study period; six were excluded, including four who had FNA of a lymph node or soft tissue lesion subsequently found to be recurrent cancer, and two who had no additional clinical information available. Thus, the study cohort included 70 nodules with TERT promoter mutations in 70 patients, including 18 (26%) with isolated TERT (iTERT) and 52 (74%) with TERT and ≥1 concurrently detected mutation (TERT+). The mean patient age in the iTERT cohort was higher than for the TERT+ (75.3 vs. 68.3 years, p = 0.007) (Table 1). Mean nodule size in the iTERT group was 2.8 cm compared to 3.3 cm in the TERT+ group (p = 0.47). FNA results were more likely to be classified as Bethesda III in the iTERT group (56%), while among the TERT+, both Bethesda IV (31%) and VI (33%) were often seen (p = 0.01). The C228T was more common at an equivalent rate in both cohorts (84%, p = 1).

Table 1.

Patient Demographics and Preoperative Characteristics of the Entire Cohort (n = 70)

	All(n = 70)	Isolated TERT (n = 18)	TERT+(n = 52)	p-Value
Mean age, years (range)	70.1 (46–89)	75.3 (53–89)	68.3 (46–87)	0.007
Female sex, n (%)	42 (58%)	10 (56%)	32 (59%)	0.79
Mean nodule size, cm (range)	3.2 (0.9–10)	2.7 (1.2–5.9)	3.4 (0.9–10)	0.18
FNA result, Bethesda Category
3	18 (25%)	10 (56%)	8 (15%)	0.01
4
FN	21 (31%)	5 (28%)	16 (31%)
HCN	6 (8%)	2 (11%)	4 (7%)
5	7 (11%)	1 (6%)	6 (13%)
6	18 (25%)	0	18 (33%)
TERT mutation identified				1
C228T	59 (84%)	15 (83%)	44 (85%)
C250T	11 (16%)	3 (17%)	8 (15%)

FN, follicular neoplasm; FNA, fine needle aspiration; HCN, Hürthle cell neoplasm.

By definition, no additional mutations were found via TSv3 testing in the iTERT group. In comparison, the most common mutations identified with TERT in the TERT+ group were BRAF^V600E (23, 44%), RAS (19, 37%), and copy number alterations (CNA; 15, 29%). Two nodules had concurrent EIF1AX alterations only in conjunction with TERT.

The mean allelic frequency (AF) of the TERT promoter mutation was 29% (range 1–75%) across all nodules and was higher in the TERT+ (32% vs. 20% p = 0.01). A total of 7/70 (8.5%) had AF <0.05, including six in the iTERT group, and none have undergone surgery. The remaining patient had TERT with BRAF^V600E and had total thyroidectomy for multifocal papillary thyroid cancer (PTC), including a 1.8-cm PTC with lymphatic invasion, extrathyroidal extension, and positive margins. Following radioactive iodine ablation (RAI), she remains without evidence of disease (NED) at 76.1 months follow-up.

Histopathology (core biopsy, lymph node biopsy, or thyroidectomy) was available for 9/18 (50%) iTERT and 51/52 (98%) TERT+ nodules (Table 2, Fig. 1). Malignancy (excluding NIFTP) was more commonly diagnosed with TERT+ as compared with iTERT (96% vs. 67%, p = 0.02). Additionally, there was a 77-year-old woman who had an iTERT (AF 40%) AUS TIRAD3 nodule, which was histologically classified as an encapsulated thyroid tumor of follicular cell origin with high-grade features. After multidisciplinary review, completion thyroidectomy and RAI ablation were administered, and the patient remains NED after 15 months follow-up. In the TERT+ group, 1 NIFTP was diagnosed in an 81-year-old woman with a 5.5-cm follicular neoplasm and NRAS and TERT detected preoperatively. After excluding the patient with encapsulated thyroid tumor of follicular origin with high-grade features, preoperative factors to assess risk of malignancy with iTERT was assessed, and there were no differences in sex (p = 0.46), Bethesda category (p = 1), type of TERT mutation (p = 1.0), nodule size (mean benign 3.7 ± 3.1 cm vs. 2.9 ± 1.0 cm, p = 0.53), or TERT AF (mean benign 29% ± 1 vs. 26% ± 1, p = 0.73).

Table 2.

Histopathology of Nodules (by Core Biopsy, Lymph Node Biopsy, or Thyroidectomy)

	Isolated TERT (n = 9)	TERT+ (n = 51)
Benign (%)	2 (22%)	1 (2%)
NIFTP (%)	0	1 (2%)
Other (%)	1 (11%)	0
Malignant (%)	6 (67%)	49 (96%)	p = 0.02
Papillary	0	27 (56%)
Follicular	1 (17%)	2 (4%)
Oncocytic	3 (50%)	4 (8%)
Poorly differentiated	2 (33%)	10 (20%)
Anaplastic	0	6 (12%)
ATA risk for DTC
Low	4 (100%)	10 (30%)
Intermediate	0	12 (35%)
High	0	12 (35%)

ATA, American Thyroid Association; DTC, differentiated thyroid carcinoma; NIFTP, noninvasive follicular thyroid neoplasm with papillary-like nuclear features.

FIG. 1.

Oncoplot of nodules with available histology (n = 60).

Malignancies in the iTERT cohort included two PDTC, one follicular thyroid cancer (FTC), and three oncocytic carcinoma (i.e., Hürthle cell carcinoma) (Table 3). All four of the differentiated thyroid cancer were minimally invasive with capsular invasion only. In contrast, the TERT+ malignancies included 6 ATC and 10 PDTC. PDTC/ATC was more likely when TERT was coexistent with ≥2 concurrently detected mutation compared with TERT with one concurrent mutation (59%, 10/17 vs. 19%, 6/32, p = 0.009) but equally likely when cytology was Bethesda III/IV compared with Bethesda V/VI (7/28, 25% vs. 8/23, 35%, p = 0.54). Among DTC with TERT+, 10/33 (30%) were histologically classified as ATA-Low risk, while the remaining 70% were ATA-Intermediate or High risk.¹³ Postoperative RAI was administered to 6/9 (67%) patients with iTERT and 35/51 (69%) with TERT+.

Table 3.

Disease-Specific Outcomes and Mean Follow-Up of Surgical Cohort (n = 55)

	Isolated TERT (n = 6)	TERT+ (n = 49)	p-Value
Mean follow-up, month (range)	37.5 (6.6–77)	27.8 (0.2–82)	0.23
Local recurrence, No./No. resectable, %	0	11/44 (25%)	0.31
Distant metastasis, n (%)	1 (17%)	16 (33%)	0.65
Deceased, n (%)	0	8 (16%)	0.57

In post hoc analysis of the four nodules with TERT which were not malignant (i.e., either benign or NIFTP), slides were able to be retrieved for three; two were confirmed benign, and one nodule had nuclear features of PTC and a focus suspicious for capsular invasion on rereview. The patient has since succumbed to nonthyroid-related disease and thus the change in diagnosis does not impact clinical care. Repeat sequencing of surgically removed tissue confirmed TERT in all three of the cases. Thus, with post hoc analysis, 98% of TERT+ nodules and 67% of iTERT nodules were malignant.

Median follow-up was 13.1 months (interquartile range 26.2, 7.2–33.4) and was not significantly different between cohorts (iTERT, 10.6 months, median IQR 46.0 [5.75–51.7] vs. 13.4, IQR 24.5 [8.5–33.1]; p = 0.37). Among patients who had initial thyroidectomy, local recurrence in the thyroid bed, mediastinum, or lateral neck has occurred in 11/44 (25%) TERT+ patients versus none in the iTERT (p = 0.31). Distant metastases were diagnosed in 17 patients (1, 17%, iTERT vs. 16, 33%, TERT+, p = 0.65). Kaplan–Meier analysis demonstrates seemingly reduced disease-free survival with TERT+ compared to iTERT, but this was not statistically significant (p = 0.11) (Fig. 2). Nine patients have died, and all were TERT+ (p = 0.57).

FIG. 2.

Disease-free survival. Kaplan–Meier survival analysis demonstrating disease-free survival (in months) for patients on the basis of preoperative detection of TERT in isolation and TERT detection in combination with other mutations (TERT+). DFS, disease-free survival.

Among patients with DTC, all 14 with ATA-Low risk disease have no biochemical or imaging evidence of disease at follow-up to date. One patient with ATA-Intermediate disease has developed locoregional recurrence, and another has elevated thyroglobulin levels after thyroidectomy and RAI with negative imaging. In the 12 patients with ATA-High risk histology, three patients have developed locoregional recurrence, four have distant metastasis, and another two have detectable thyroglobulin after initial treatment without evidence of structural disease. Thus, in the subset of DTC patients with TERT mutations, recurrent/persistent disease is present in 0%, 18%, and 75% in ATA-Low, IM, and High-Risk disease categories, respectively.

In surveillance of 10 nodules which were not resected (nine with iTERT and one with TERT and BRAF^V600E), none have enlarged or developed lymph node/distant metastasis at a mean follow-up of 18.5 months. Two nodules have had repeat FNA with results which were benign in one, and AUS with TERT detected at the same AF as the initial biopsy in the other. The patient with a TERT and BRAF^V600E Bethesda VI nodule also had metastatic breast cancer, and thus surgical management of her thyroid was deferred up to her demise.

Discussion

The use of molecular testing for indeterminate cytology has already been demonstrated to reduce the need for diagnostic surgery with associated cost-savings despite the added test cost.¹⁴ However, in addition to the risk of malignancy, the specific detected genetic alteration can provide preoperative information on possible type of cancer and be valuable prognostically.^4,5 While TERT has been demonstrated in many studies to be associated with aggressive cancers diagnosed after histology and surveillance, the significance of preoperatively detected TERT has remained somewhat elusive. To our knowledge, this is also the first study to report on TERT mutation identified in isolation. In this single institution series, we identified 70 nodules in 70 unselected patients who had TERT detected in a thyroid nodule being assessed for initial management; the majority were TERT+ (74%), and 92% were malignant.

Even more importantly for patient management, 18/55 (33%) malignancies with TERT were either poorly differentiated or anaplastic. Two of the poorly differentiated thyroid cancer had iTERT in patients with 2.8-cm Bethesda III and 4.4-cm Bethesda IV nodules. Additionally, of these 18 non-DTC malignancies, 5 were classified as Bethesda VI, and 3 were Bethesda V, and the detection of TERT was useful clinically to direct definitive management. Among the PDTC, some had areas of differentiated cancer with progression to PD likely driven by the presence of TERT, which is consistent with in vitro studies assessing BRAF and TERT overexpression in mice and demonstrates ∼30% progression to PDTC.¹⁵ One of these patients was a 50-year-old woman who had a 9-mm TIRADS 5 nodule who was undergoing active surveillance prior to being seen to establish continuity of care. Her biopsy was NRAS and TERT positive, and a 1-cm minimally invasive FTC was diagnosed with progression to PDTC noted histologically. Thus, nodules with TERT should be considered for either surgery or close interval surveillance. Whether these nodules with already a predisposing genetic marker for tumor dedifferentiation can safely be treated with RFA remains to be assessed.

Despite the relatively large proportion of non-DTC, the majority of malignancies (n = 37) were still differentiated, and 40% were histologically ATA-Low risk. On a relatively short follow-up, none of these have recurred or developed distant metastasis. Whether the presence of TERT is an independent risk factor for recurrence and/or metastatic disease when histology is seemingly indolent remains to be seen in future studies with longer follow-up. Distant metastasis from cancers with low-risk histology and TERT + RAS molecular profiles as well as late metastases from TERT+ tumors have been documented.^16–18 In this study, even in short-term follow-up, structural locoregional recurrence has been diagnosed in one ATA-Intermediate PTC patient with TERT and BRAF^V600E and in three patients with ATA-High-risk DTCs. Additionally, four patients with ATA-High-risk DTC have distant metastasis. In a recent study of 171 patients with PTC surgically resected, the presence of TERT increased the likelihood of distant metastasis (odds ratio [OR] 51.3, 7–1048.2, p < 0.001) and coexisting TERT and BRAF increased the risk of radioiodine-refractory disease (OR 21.7, 5.6–88.9, p < 0.001).¹⁹ We were unable to observe any association between locoregional recurrence and distant metastasis with coexisting mutation. Preoperative detection of TERT, particularly in conjunction with another driver mutation, should raise concern for not just malignancy but a higher than baseline risk for locoregional recurrence and/or distant metastasis, and total thyroidectomy if feasible should be considered to, at minimum, facilitate surveillance.

Not all TERT nodules were malignant, as previously described.¹² We observed three benign nodules (two adenomas and one hyperplastic nodule) and one NIFTP. Three of these patients additionally had PTMC or microscopic NIFTP in the resected thyroid. In this relatively small cohort, we were unable to identify specific features to help identify what risk factors were associated with malignancy. The AF of the detected TERT mutation for these nonmalignant nodules ranged from 28% to 40%, and the likelihood of false positive testing is low. Despite the detection of preoperative TERT with or without concurrent driver mutation, patients were managed according to final histology.

Surveillance has been performed for 10 nodules: nine with iTERT and one in the TERT+ cohort in a patient with metastatic breast cancer who has since died. In the nine patients with iTERT, five are age >80 years and all have nodule size <4 cm. While the exact reason for deferred surgical management is not known, likely these clinical variables were a contributing reason. In addition, six of these nodules had TERT mutation at a very low level (allele frequency of <5%, with predicted low probability of cancer on molecular testing), and to date, the nodules have not required resection. While the presence of an iTERT is associated with a high risk of malignancy, as evidenced from the histological results in the present study, not all are aggressive on limited follow up and certainly if patients are not candidates for surgery then continued imaging surveillance may be considered.

The limitations of this study include short follow-up and a small sample size, as TERT mutations both in isolation and with additional driver mutations are a relatively uncommon event in thyroid cancer, and molecular testing has only more recently been widely used. We do not have details on the number of nodules during this same time frame to know what proportion were TERT positive, although in a different study of 50,734 FNA specimens tested with TSv3, TERT mutations were detected in 5.1% of nodules with Bethesda III–BVI cytology.⁹ Indication for molecular testing was in the context of clinical management, and particularly in the Bethesda VI category, there was potentially a selection bias toward more aggressive tumors being considered for systemic therapy, which may in part explain the high proportion (one-third) of PDTC/ATC among cases. The present study was also a single-institution study that was unable to account for any geographic variations. Additionally, the pathologists were not blinded to the cytological or molecular results, which could produce bias. Lastly, not all nodules had histology, and thus the malignancy rate may be lower than reported.

In summary, a TERT promoter mutation detected on preoperative molecular testing is frequently co-occurring with additional oncogenic drivers or structural genomic alterations, such as CNAs. However, the presence of TERT, whether in isolation or in conjunction with other mutations, is associated with a high risk of malignancy, at 93%. Notably, up to one-third of these cases exhibit high-grade histology, including poorly differentiated or anaplastic thyroid carcinoma, and this risk appears comparable regardless of whether the TERT mutation occurs alone or alongside other molecular alterations. A larger series of cases with long-term follow-up is needed to conclude if DTC with iTERT mutations can be managed more conservatively. Surgical management favoring total thyroidectomy should be strongly considered for nodules with TERT coexisting with other alterations since even on short follow-up, these are associated with local recurrence and distant metastasis.

Authors’ Contributions

M.C.: Writing—review and editing, writing—original draft, visualization, validation, methodology, investigation, formal analysis, data curation, and conceptualization. A.S.: Writing—review and editing, conceptualization, and investigation. E.K.-F.: Writing—review and editing and conceptualization. E.M.M.: Writing—review and editing and conceptualization. K.L.M.: Writing—review and editing, investigation, and conceptualization. K.M.R.: Writing—review and editing, investigation, conceptualization, and methodology. S.K.: Writing—review and editing and conceptualization. N.P.O.: Writing—review and editing and conceptualization. S.I.C.: Writing—review and editing and conceptualization. R.R.S.: Writing—review and editing and conceptualization. S.E.C.: Writing—review and editing and conceptualization. M.N.N.: Writing—review and editing, conceptualization, and methodology. Y.E.N.: Writing—review and editing, investigation, formal analysis, conceptualization, resources, and methodology. L.Y.: Writing—review and editing, writing—original draft, supervision, methodology, conceptualization, formal analysis, visualization, and data curation.

Author Disclosure Statement

M.N.N. and Y.E.N. own intellectual property related to ThyroSeq® and receive royalties through the University of Pittsburgh; they also serve as consultants for Sonic Healthcare USA. SIC serves as a consultant for Sonic Healthcare USA. The other authors have no conflicts of interest to report.

Funding Information

M.C. received a T35 National Institutes of Health training grant for this work (T35DK065521). S.E.C. was supported by a generous gift from the William and Susan Johnson Fund for Endocrine Surgery Research for work unrelated to this.