Skip to main content
NCBI home page
European Thyroid Journal logoLink to European Thyroid Journal
. 2025 Dec 9;14(6):e250296. doi: 10.1530/ETJ-25-0296

Comparing the diagnostic accuracy of Afirma GSC to ThyroSeq V3 in cytologically indeterminate thyroid nodules

Natasha Dowell 1,✉, Shayma Begum 1, Jameel Muzaffar 2, Kristien Boelaert 3, Hannah Nieto 4
PMCID: PMC12697243  PMID: 41324437

Abstract

Objective

To compare the diagnostic test accuracy of Afirma GSC and ThyroSeq v3 in cytologically indeterminate thyroid nodules.

Methods

PubMed, Embase, Cochrane Library, Medline, and the Web of Science were searched from the date of inception to May 9, 2025. Two independent reviewers screened articles for eligibility. Studies assessing Afirma gene sequencing classifier (GSC) or ThyroSeq v3 for indeterminate thyroid nodules were included. The Preferred Reporting Items for Systematic Reviews and Meta-Analysis reporting guidelines were followed. Statistical analysis was performed using R.

Results

A total of 26 studies met the eligibility criteria. For surgically confirmed Afirma GSC results, the sensitivity, specificity, and negative predictive value (NPV) were 94, 42, and 96%, respectively. For unoperated negative cases, the values were 96, 86, and 99%. For ThyroSeq v3, the sensitivity, specificity, and NPV for surgically confirmed cases were 96, 40, and 93%, respectively, and for unoperated negative cases, they were 97, 83, and 99%.

Conclusion

Both molecular tests demonstrate high NPV but low specificity; neither is clearly superior. Future research should prioritise randomised controlled trials, long-term follow-up of unoperated nodules, and direct comparisons of molecular tests.

Keywords: thyroid, thyroid nodule, thyroid cancer

Introduction

Thyroid nodules are estimated to be present in 6–67% of the population, and while most are benign, 10–15% are found to be malignant (1, 2). Fine-needle aspiration cytology (FNAC) is the standard investigation for classifying thyroid nodules as benign or malignant, providing management guidance (3). A limitation of FNAC is that an estimated 25% of patients will receive a cytologically indeterminate result, meaning it is uncertain if the nodule is benign or malignant (4). Indeterminate thyroid nodules include Bethesda III (atypia of undetermined significance or follicular lesion of undetermined significance) and IV (follicular neoplasm or suspicious for follicular neoplasm) (5). In most indeterminate cases, the patient will undergo diagnostic hemithyroidectomy so the nodule can be examined histologically for a final diagnosis. The malignancy rate of operated indeterminate nodules is reported at 33.9%, meaning many benign nodules are operated on (6). This poses a significant burden on healthcare systems through unnecessary surgeries, which can also negatively impact patients’ quality of life.

The American Thyroid Association recommends molecular testing to guide the management of cytologically indeterminate thyroid nodules (7). These tests aim to reduce unnecessary diagnostic surgeries and optimise patient outcomes. There are several commercially available molecular tests. Afirma Genomic Sequencing Classifier (GSC), (developed by Veracyte Inc, USA) analyses mRNA expression to determine malignant potential, and represents a development from the previous generation, Afirma gene expression classifier (GEC) (8). ThyroSeq v3 genomic classifier uses next-generation sequencing to assess 112 gene alterations associated with thyroid carcinoma (9). Both tests report samples as benign or suspicious.

Previous meta-analyses have primarily focused on individual molecular tests or earlier test generations (10, 11, 12, 13, 14). Two previous reviews compared the diagnostic accuracy between ThyroSeq v3, Afirma GSC, and microRNA panels. Both reviews had a strong methodology; however, they were both limited by their sample size (15, 16).

A previous review assessed the real-world performance of Afirma GSC by assuming unoperated molecular test negative cases are true negatives. This was because most patients in these studies did not have surgery (and final histology), with a negative GSC result, but instead had follow-up on ultrasound for 6–12 months assessing nodule growth (11). No previous review has analysed the performance of ThyroSeq v3 in this way. Our review assesses real-world performance of Afirma GSC and ThyroSeq v3 by analysing diagnostic performance in both surgically confirmed nodules and unoperated molecular test negative (assumed true negative) nodules. This approach provides a more comprehensive understanding of the diagnostic performance of these molecular tests and addresses gaps in the existing literature.

Materials and methods

This systematic review and meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (17). Before the literature screening, this review was prospectively registered through the International Prospective Register of Systematic Reviews (PROSPERO) (Registration number: CRD42024578531).

Literature search and data collection

A description of the participants, interventions, comparisons, outcomes, timing, and study design is provided in Table 1. The following databases were searched from date of inception to May 9, 2025: PubMed, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), OVID MEDLINE, and the Web of Science. The search terms were designed with the support of a research librarian (Table 2). Additional texts were identified through reference searching of the included studies. Articles were screened for eligibility by two independent reviewers (ND, SB). Any discrepancies were resolved through discussion with a third independent reviewer (HN).

Table 1.

Population, interventions, comparisons, outcomes, timing, study design (PICOTS).

PICOTS Description
Population All patients (no age limitation) with cytologically indeterminate thyroid nodules where the sample was adequate but not possible to differentiate between benign and malignant disease
Interventions Afirma GSC or ThyroSeq version 3 with histopathologic diagnosis
Comparisons Each other
Outcomes Diagnostic accuracy of the molecular test
Timing Any time point
Study design A systematic review and meta-analysis
Open in a new tab

Table 2.

PubMed search terms.

PubMed (‘Indeterminate’[title/abstract] OR ‘Thy3’[title/abstract] OR ‘Thy4’[title/abstract] OR ‘bethesda category 3’[title/abstract] OR ‘bethesda category 4’[title/abstract] OR ‘bethesda category iii’[title/abstract] OR ‘bethesda category iv’[title/abstract]) AND (‘ThyroSeq’[title/abstract] OR ‘Afirma’[title/abstract] OR ‘Affirma’[title/abstract] OR ‘GSC’[title/abstract] OR ‘molecular diagnostic’[title/abstract] OR ‘genetic testing’[title/abstract] OR ‘molecular marker’[title/abstract] OR ‘genom*’[title/abstract] OR ‘ThyGenX’[title/abstract] OR ‘ThyGeNEXT’[title/abstract] OR ‘ThyraMIR’[title/abstract] OR ‘Rosetta’[title/abstract] OR ‘ThyroSpec’[title/abstract] OR ‘miRInform’[title/abstract] OR ‘AmpliSeq’[title/abstract] OR ‘mir-THYpe’[title/abstract] OR ‘gene sequencing classifier’[title/abstract] OR ‘next generation sequencing’[title/abstract] OR ‘cytopathology’[title/abstract]) AND (‘thyroid gland’[title/abstract] OR ‘thyroid nodule’[title/abstract] OR ‘thyroid neoplasms’[title/abstract] OR ‘thyroid cancer’[title/abstract] OR ‘thyroid carcinoma’[title/abstract] OR ‘thyroid microcarcinoma’[title/abstract] OR ‘thyroid tum*’[title/abstract] OR ‘thyroid adenoma’[title/abstract] OR ‘thyroid adenocarcinoma’[title/abstract] OR ‘thyroid nod*’[title/abstract] OR ‘thyroid lump’[title/abstract] OR ‘thyroid swelling’[title/abstract] OR ‘follicular’[title/abstract] OR ‘papillary’[title/abstract] OR ‘malignant’[title/abstract]) AND (‘biopsy, fine needle’[MeSH terms] OR (‘biopsy’[all fields] AND ‘fine needle’[all fields]) OR ‘fine-needle biopsy’[all fields] OR (‘fine’[all fields] AND ‘needle’[all fields] AND ‘aspiration’[all fields]) OR ‘fine needle aspiration’[all fields] OR (‘biopsy, fine needle’[MeSH terms] OR (‘biopsy’[all fields] AND ‘fine needle’[all fields]) OR ‘fine-needle biopsy’[all fields] OR (‘fine’[all fields] AND ‘needle’[all fields] AND ‘biopsy’[all fields]) OR ‘fine needle biopsy’[all fields]))
Open in a new tab

Eligibility criteria

Inclusion criteria:

  • –

    All patients (no age limitation) with cytologically indeterminate thyroid nodules (Bethesda III or IV).

  • –

    Articles must evaluate either ThyroSeq V3 or Afirma GSC as an index test.

  • –

    The article must contain true positive (TP), false positive (FP), false negative (FN), and true negative (TN) results to calculate the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

  • –

    All patients must have histopathological confirmation.

Exclusion criteria:

  • –

    Duplicates, review articles, comments, conference abstracts, case reports, and unpublished articles.

  • –

    Studies with missing data regarding TP, FP, TN, or FN.

  • –

    Older-generation molecular tests or next generation sequencing platforms, for example, but not limited to: Afirma GEC, ThyroSeq v2, ThyraMIR, ThyGeNEXT, or multiplatform tests.

  • –

    Studies not published in the English language.

Non-invasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) was categorised as malignant by most included studies, and therefore this review classified them as such, despite these lesions not being considered malignant (16). There was no limitation on year of publication. The reference lists of all included articles and any relevant systematic reviews were screened by the reviewers.

Older-generation molecular tests, such as Afirma GEC and ThyroSeq v2, were excluded from this review, as our aim was to evaluate the most current versions to provide a true update to the existing literature. These earlier models have already been extensively discussed in prior reviews. Similarly, newly emerging next-generation sequencing and multiplatform tests were also excluded due to limited data availability and lack of direct comparability.

Data extraction

Two independent reviewers (ND, SB) extracted the relevant data using a predefined Excel (Microsoft, USA) table. Any discrepancies were resolved by discussion with a third reviewer (HN). The following data were extracted from each article:

  • –

    Study characteristics: first author, publication year, country.

  • –

    Patient characteristics: participant number, patient selection method, mean age, male:female ratio.

  • –

    Outcome data: TPs, FPs, TNs, and FNs, sensitivity, specificity, PPV, NPV.

Study quality

Two independent reviewers (ND and SB) assessed the studies’ methodological quality using the Downs and Black checklist (18). Studies were categorised as excellent, good, fair, or poor quality.

Statistical analysis

For each article, the TP, FP, TN, and FN were used to calculate the sensitivity, specificity, PPV, and NPV. Sensitivity and specificity, with their 95% confidence intervals, were pooled using a bivariate random-effects model. All statistical analysis was performed with R version 4.4.3 using the packages ‘Meta’, ‘Mada’, ‘ggplot2’, ‘pROC’, and ‘Metafor’ (R Project for Statistical Computing, Vienna, Austria) (19, 20, 21, 22, 23, 24). Study heterogeneity was quantified using the I2 statistic. P values ≤0.05 were considered statistically significant. Studies with nodules with a positive molecular test result that did not undergo surgery were excluded from the calculations. Studies with nodules with a negative molecular test that did not undergo surgery were recorded separately for comparison.

Results

Study characteristics

Initial searches yielded 1,866 studies (Fig. 1). Eight hundred duplicates were removed, and 1,066 were screened at the title and abstract stage, leaving 224 for full-text review. Following screening, 26 studies identified through database searches and two identified through citation searching were included in the review. Eight studies were multi-centre, and 18 were based at a single centre. Patient and study characteristics are provided in eTables 2 and 3 (see section on Supplementary materials given at the end of the article). Publication years ranged from 2015 to 2025. The mean age of patients who had Afirma GSC and ThyroSeq v3 testing was 53.6 and 54.7 years, respectively. A total of 1,127 and 1,162 surgically confirmed nodules were assessed using ThyroSeq v3 and Afirma GSC, respectively. Twenty-one studies were based in the USA, one study was based in the USA and Singapore, one study from the Netherlands, one from Singapore, and one from Canada. One study was a randomised controlled trial, 7 were prospective, and 18 contained historical data.

Figure 1.

Figure 1

Open in a new tab

Preferred reporting items for systematic reviews and meta-analyses (PRISMA) flowchart.

Quality and publication bias

The quality assessment scores using the Downs and Black checklist are provided in eTable 4. The scores ranged from 12 to 22 out of a total of 28. Five studies were categorised as good quality, 17 as fair quality, and four were of poor quality. The main reason behind the range in quality assessment was the lack of randomisation across studies.

Results regarding the diagnostic performance of the molecular panels

Afirma GSC: operated nodules with histological diagnosis

Sixteen studies, including 1,162 surgically confirmed nodules, evaluated Afirma GSC (5, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39). Surgically confirmed cases had a range of sensitivities from 0.91 to 1.00, eTable 5. Specificity ranged from 0.17 to 0.68. Within this surgically confirmed Afirma GSC subgroup, the pooled estimates of sensitivity and specificity were 0.94 (95% CI: 0.91–0.97) and 0.42 (95% CI: 0.27–0.59), respectively, as shown in Fig. 2. The negative predictive value (NPV) was 0.96 (95% CI: 0.89–0.98). Pooled summary data is provided in Table 3.

Figure 2.

Figure 2

Open in a new tab

Forest plot showing the sensitivity (A) and specificity (B) for Afirma GSC, where only surgically confirmed cases are included, and sensitivity (C) and specificity (D) assuming all unoperated test-negative cases are true negatives. Squares represent the point estimate for an individual study. The line represents the 95% confidence interval. The red square represents the overall pooled effect.

Table 3.

Pooled summary data. The 95% CI values are provided in parentheses.

SC-Afirma GSC Afirma GSC* SC-ThyroSeq v3 ThyroSeq v3*
Sensitivity 0.96 (0.91–0.98) 0.98 (0.92–0.99) 0.96 (0.94–0.97) 0.97 (0.92–0.99)
Specificity 0.39 (0.32–0.47) 0.84 (0.79–0.87) 0.40 (0.33–0.47) 0.83 (0.79–0.87)
NPV 0.96 (0.89–0.99) 1.00 (0.98–1.00) 0.93 (0.85–0.96) 0.99 (0.98–1.00)
Open in a new tab
*

Assuming all test-negative unoperated cases are true negatives.

NPV, negative predicted value; SC, surgically confirmed.

Afirma GSC: non-operated nodules with assumed negative diagnosis

Sixteen studies reported test-negative cases that did not undergo surgical intervention, eTable 6, (5, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39). Assuming all test-negative cases that were not operated on were true-negatives, the sensitivity ranged from 0.63 to 1.00, and specificity ranged from 0.71 to 0.95. The pooled estimates of sensitivity and specificity were 0.96 (95% CI: 0.92–0.98) and 0.86 (95% CI: 0.81–0.90), respectively, Fig. 2. The negative predictive value was 0.99 (95% CI: 0.98–1.00), which was expected when assuming all unoperated cases are true negatives, Fig. 3.

Figure 3.

Figure 3

Open in a new tab

Forest plot showing the NPV for Afirma GSC as a molecular test, where only surgically confirmed cases are included (A), and assuming all unoperated test-negative cases are true negatives (B). Squares represent the point estimate for an individual study. The line represents the 95% confidence interval. The red square represents the overall pooled effect.

Loncar et al. did not specify the number of surgically confirmed true negatives in their GSC cohort; they assumed that a negative ultrasound at 6 months meant the nodule was a true negative, and therefore its results were not included in the surgically confirmed cases (39).

ThyroSeq V3 next-generation sequencing: operated nodules with histological diagnosis

Thirteen studies, comprising 1,127 surgically confirmed nodules, assessed the performance of ThyroSeq V3 (5, 9, 26, 34, 40, 41, 42, 43, 44, 45, 46, 47, 48). Cases with confirmed surgical follow-up demonstrated a range of sensitivities from 0.80 to 1.00, eTable 7. Specificity ranged from 0.03 to 1.00. Within this surgically confirmed subgroup, the pooled estimates of sensitivity and specificity were 0.96 (95% CI: 0.94–0.97) and 0.40 (95% CI: 0.33–0.47) respectively, Fig. 4. The negative predictive value was calculated as 0.93 (95% CI: 0.85–0.96), Fig. 5.

Figure 4.

Figure 4

Open in a new tab

Forest plot showing the sensitivity (A) and specificity (B) for ThyroSeq v3, where only surgically confirmed cases are included, and sensitivity (C) and specificity (D) assuming all unoperated test-negative cases are true negatives. Squares represent the point estimate for an individual study. The line represents the 95% confidence interval. The red square represents the overall pooled effect.

Figure 5.

Figure 5

Open in a new tab

Forest plot showing the NPV for ThyroSeq v3, where only surgically confirmed cases are included (A), and assuming all unoperated test-negative cases are true negatives (B). Squares represent the point estimate for an individual study. The line represents the 95% confidence interval. The red square represents the overall pooled effect.

ThyroSeq V3 next generation sequencing: non-operated nodules with assumed negative diagnosis

Nine studies reported test-negative cases that were not operated on (5, 26, 34, 40, 42, 43, 44, 45, 46, 48). Assuming all unoperated test-negative cases were true negatives, the sensitivities ranged from 0.8 to 1.00, eTable 8. Specificity ranged from 0.76 to 1.00. The pooled estimates of sensitivity and specificity were 0.97 (95% CI: 0.92–0.99) and 0.83 (95% CI: 0.79–0.87), respectively, Fig. 4. The negative predictive value was 0.99 (95% CI: 0.98–1.00), which was as expected when assuming all unoperated cases are true negatives, Fig. 5.

Comparing Afirma GSC and GEC

Nine studies reported surgically confirmed data for both Afirma GSC and GEC (27, 28, 29, 30, 31, 32, 33, 34, 37), eTable 9. A random effects model compared the sensitivity and specificity between the two groups, as seen in Fig. 6. There was a significant difference, indicating that GSC is superior: sensitivity (relative risk: 1.69 (95% CI: 1.40–2.05), specificity (relative risk: 2.95 (95% CI: 2.11–4.13)).

Figure 6.

Figure 6

Open in a new tab

Meta-analysis comparing the sensitivity (A) and specificity (B) of surgically confirmed cases between Afirma GEC and GEC. Squares represent the point estimate for an individual study. The line represents the 95% confidence interval. The red square represents the overall pooled effect.

Discussion

This review assessed the diagnostic accuracy of the Afirma GSC and ThyroSeq v3 molecular tests for detecting thyroid carcinoma in patients with cytologically indeterminate thyroid nodules. The management of indeterminate thyroid nodules poses a clinical challenge, and molecular testing reduces unnecessary diagnostic surgeries.

Key findings

This review provides an updated synthesis of the literature, including 13 ThyroSeq v3 and 16 Afirma GSC studies, which reflects a larger sample size compared to prior meta-analyses. This review demonstrated that both Afirma GSC and ThyroSeq v3 exhibit high sensitivity (GSC: 0.94 (95% CI: 0.91–0.97); ThyroSeq v3: 0.96 (95% CI: 0.94–0.97)) and a high NPV (GSC: 0.96 (95% CI: 0.89–0.98); ThyroSeq v3: 0.93 (95% CI: 0.85–0.96)). These results demonstrate the value of both tests in ruling out malignancy and reducing unnecessary surgeries. However, both tests have low specificity (Afirma GSC: 0.42 (95% CI: 0.27–0.59); ThyroSeq v3: 0.40 (95% CI: 0.33–0.47)), limiting their reliability in ruling in malignancy. This result is expected, as most molecular tests were developed for exclusion of malignancy, most easily assessed here by NPV. A previous review by Lee et al., analysed six ThyroSeq v3 and seven Afirma GSC studies (16). Their reported sensitivities (0.95 for ThyroSeq V3; 0.96 for Afirma GSC) and specificities (0.50 for ThyroSeq V3; 0.53 for Afirma GSC) align with the findings of this review. Our conclusions are further supported by the comprehensive meta-analysis from Vardali et al., which reported a sensitivity for Afirma GSC and ThyroSeq v3 of 0.97 (0.91–0.99) and 0.95 (0.91–0.97) respectively (49). Their paper reported a specificity for Afirma GSC of 0.50 (0.40–0.61) and for ThyroSeq V3 of 0.47 (0.22–0.73) (49). Our review provides a broader analysis by assessing both surgically confirmed and unconfirmed cases separately, as well as providing an update on the published literature.

Analysis of unoperated cases (assumed true negatives)

Most benign nodules identified by molecular testing are monitored through clinical follow-up with ultrasound, meaning that relying on surgically confirmed benign nodules does not provide an accurate representation of real-life performance (50). Drawing conclusions based on a small number of test-negative nodules that underwent surgery introduces selection bias. Therefore, it was important for this review to also consider data from cases that remained stable at clinical follow-up. Under this approach, sensitivity increased marginally, but specificity had a greater rise: 0.86 (95% CI: 0.81–0.90) for Afirma GSC and 0.83 (95% CI: 0.79–0.87) for ThyroSeq v3, NPV reached 0.99 (95% CI: 0.98–1.00) and 0.99 (95% CI: 0.98–1.00), respectively. A limitation of this method was the variation in clinical follow-up duration across studies. It is important to note that these values are based on assumption, which has likely resulted in inflation, and the results should be interpreted cautiously. A previous review by Nasr et al. applied the same approach for Afirma GSC alone, whereas our review is the first to document this for ThyroSeq v3 (11). In their review, Nasr et al. assessed the diagnostic accuracy of Afirma GSC and reported a sensitivity of 0.97 and a negative predictive value (NPV) of 0.995, which closely aligns with the findings for Afirma GSC in our review (11).

Limitations

The primary limitation of this review is the variability in methodology and reporting across included studies. There was only one randomised controlled trial identified, and there was a lack of direct comparisons of molecular tests within the same cohort, highlighting a need for more robust research.

There is also limited data investigating the impact of molecular test results on quality of life (QoL). Our review identified one study examining Afirma GSC and ThyroSeq v3, drawn from the same cohort as the only randomised controlled trial included (26, 51). Their study reported sustained QoL in patients with a ‘benign’ result, with a poorer QoL in the ‘suspicious’ group, which improved postoperatively. There was no comparison of pre- and post-test QoL. Further studies should focus on the effects of false negatives and false positives on quality of life, and the impact on QoL of avoiding surgery in patients with ‘benign’ results.

A key limitation of this paper and the studies it includes is that the true sensitivity of molecular tests cannot be determined. Most nodules classified as ‘benign’ by a molecular test are not surgically removed and therefore lack histopathological results. Consequently, the observed sensitivity in surgically confirmed cohorts is likely to be overestimated.

Conclusion

Both Afirma GSC and ThyroSeq v3 are effective tools for managing indeterminate thyroid nodules, primarily by ruling out malignancy and reducing unnecessary diagnostic surgeries. Limited evidence supports the superiority of one test over the other, and both remain appropriate diagnostic tests. Future research should prioritise randomised controlled trials, long-term follow-up of unoperated nodules, and direct comparisons of molecular tests to assess the impact on patients.

Supplementary materials

Declaration of interest

There is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding

This research did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.

Author contribution statement

ND contributed to the design of the work, first reviewer, literature searches, data extraction, and writing of the paper. SB was responsible for literature searches, second reviewer, and data extraction. JM helped in revision of text, critical appraisal, and helped draft and edit writing. KB critically appraised data and writing, and helped draft writing. HN supervised the project, critically appraised data, revised text, and suggested the study question.

References

  • 1.Dean DS & Gharib H. Epidemiology of thyroid nodules. Best Pract Res Clin Endocrinol Metabol 2008. 22 901–911. ( 10.1016/j.beem.2008.09.019) [DOI] [PubMed] [Google Scholar]
  • 2.Kamran SC, Marqusee E, Kim MI, et al. Thyroid nodule size and prediction of cancer. J Clin Endocrinol Metab 2013. 98 564–570. ( 10.1210/jc.2012-2968) [DOI] [PubMed] [Google Scholar]
  • 3.Langer JE, Baloch ZW, McGrath C, et al. Thyroid nodule fine-needle aspiration. Seminars Ultrasound CT MRI 2012. 33 158–165. ( 10.1053/j.sult.2011.12.002) [DOI] [PubMed] [Google Scholar]
  • 4.Bongiovanni M, Spitale A, Faquin WC, et al. The Bethesda system for reporting thyroid cytopathology: a meta-analysis. Acta Cytol 2012. 56 333–339. ( 10.1159/000339959) [DOI] [PubMed] [Google Scholar]
  • 5.Kim NE, Raghunathan RS, Hughes EG, et al. Bethesda III and IV thyroid nodules managed nonoperatively after molecular testing with Afirma GSC or Thyroseq v3. J Clin Endocrinol Metab 2023. 108 e698–e703. ( 10.1210/clinem/dgad181) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Dimitriadis P, Moinie A, Michaels J, et al. Indeterminate thyroid nodules (Thy3): malignancy rate and characteristics in a study of 118 patients. Ann R Coll Surg Engl 2023. 105 568–571. ( 10.1308/rcsann.2022.0163) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Ferris RL, Baloch Z, Bernet V, et al. American Thyroid Association statement on surgical application of molecular profiling for thyroid nodules: current impact on perioperative decision making. Thyroid 2015. 25 760–768. ( 10.1089/thy.2014.0502) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Hu MI, Waguespack SG, Chrysoula D, et al. Afirma genomic sequencing classifier and xpression atlas molecular findings in consecutive Bethesda III–VI thyroid nodules. J Clin Endocrinol Metab 2021. 106 2198–2207. ( 10.1210/clinem/dgab304) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Steward DL, Carty SE, Sippel RS, et al. Performance of a multigene genomic classifier in thyroid nodules with indeterminate cytology. JAMA Oncol 2019. 5 204. ( 10.1001/jamaoncol.2018.4616) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Vuong HG, Nguyen TPX, Hassell LA, et al. Diagnostic performances of the Afirma gene sequencing classifier in comparison with the gene expression classifier: a meta‐analysis. Cancer Cytopathology 2020. 129 182–189. ( 10.1002/cncy.22332) [DOI] [PubMed] [Google Scholar]
  • 11.Nasr CE, Andriol M, Endo M, et al. Real world performance of the Afirma genomic sequencing classifier (GSC) – a meta-analysis. J Endocr Soc 2022. 6(Supp. Supplement_1) A745–A746. ( 10.1210/clinem/dgac688) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Borowczyk M, Szczepanek-Parulska E, Olejarz M, et al. Correction to: evaluation of 167 gene expression classifier (GEC) and ThyroSeq v2 diagnostic accuracy in the preoperative assessment of indeterminate thyroid nodules: bivariate/hroc meta-analysis. Endocr Pathol 2019. 30 16. ( 10.1007/s12022-019-9566-7) [DOI] [PubMed] [Google Scholar]
  • 13.Liu Y, Pan B, Xu L, et al. The diagnostic performance of Afirma gene expression classifier for the indeterminate thyroid nodules: a meta-analysis. Biomed Res Int 2019. 2019 1–11. ( 10.1155/2019/7150527) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Vargas-Salas S, Luis Martínez J, Urra S, et al. Genetic testing for indeterminate thyroid cytology: review and meta-analysis. Endocr Relat Cancer 2017. 25 R163–R177. ( 10.1530/erc-17-0405) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Silaghi CA, Lozovanu V, Georgescu CE, et al. Thyroseq v3, Afirma GSC, and microRNA panels versus previous molecular tests in the preoperative diagnosis of indeterminate thyroid nodules: a systematic review and meta-analysis. Front Endocrinol 2021. 12 649522. ( 10.3389/fendo.2021.649522) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Lee E, Terhaar S, McDaniel L, et al. Diagnostic performance of the second-generation molecular tests in the assessment of indeterminate thyroid nodules: a systematic review and meta-analysis. Am J Otolaryngol 2022. 43 103394. ( 10.1016/j.amjoto.2022.103394) [DOI] [PubMed] [Google Scholar]
  • 17.Page M, McKenzie J, Bossuyt P, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021. 372 71. ( 10.1136/bmj.n71) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Downs SH & Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 1998. 52 377–384. ( 10.1136/jech.52.6.377) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.R Core Team . R: The R Project for Statistical Computing [Internet]: R-Project.Org, 2022. (https://www.r-project.org) [Google Scholar]
  • 20.Balduzzi S, Rücker G & Schwarzer G. How to perform a meta-analysis with R: a practical tutorial. Evid Base Ment Health 2019. 22 153–160. ( 10.1136/ebmental-2019-300117) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Sousa-Pinto P. Mada: meta-analysis of diagnostic accuracy. R Package Version 0.5.11, [Internet]: R-Project.Org, 2022. [cited 2024 Dec 11]. (https://cran.r-project.org/package=mada) [Google Scholar]
  • 22.Wickham H. ggplot2: Elegant Graphics for Data Analysis. New York: Springer-Verlag, 2016. [Google Scholar]
  • 23.Robin X, Turck N, Hainard A, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinforma 2011. 12 77. ( 10.1186/1471-2105-12-77) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Viechtbauer W. Conducting meta-analyses in R with the metafor package. J Stat Software 2010. 36 1–48. ( 10.18637/jss.v036.i03) [DOI] [Google Scholar]
  • 25.Patel KN, Angell TE, Babiarz JE, et al. Performance of a genomic sequencing classifier for the preoperative diagnosis of cytologically indeterminate thyroid nodules. JAMA Surg 2018. 153 817–824. ( 10.1001/jamasurg.2018.1153) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Livhits MJ, Zhu CY, Kuo EJ, et al. Effectiveness of molecular testing techniques for diagnosis of indeterminate thyroid nodules. JAMA Oncol 2021. 7 70. ( 10.1001/jamaoncol.2020.5935) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Angell TE, Heller HT, Cibas ES, et al. Independent comparison of the Afirma genomic sequencing classifier and gene expression classifier for cytologically indeterminate thyroid nodules. Thyroid 2019. 29 650–656. ( 10.1089/thy.2018.0726) [DOI] [PubMed] [Google Scholar]
  • 28.Endo M, Nabhan F, Porter K, et al. Afirma gene sequencing classifier compared with gene expression classifier in indeterminate thyroid nodules. Thyroid 2019. 29 1115–1124. ( 10.1089/thy.2018.0733) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Harrell RM, Eyerly-Webb SA, Golding AC, et al. Statistical comparison of Afirma Gsc and Afirma Gec outcomes in a community endocrine surgical practice: early findings. Endocr Pract 2018. 25 161–164. ( 10.4158/ep-2018-0395) [DOI] [PubMed] [Google Scholar]
  • 30.San Martin VT, Lawrence L, Bena J, et al. Real-world comparison of Afirma GEC and GSC for the assessment of cytologically indeterminate thyroid nodules. J Clin Endocrinol Metab 2020. 105 dgz099. ( 10.1210/clinem/dgz099) [DOI] [PubMed] [Google Scholar]
  • 31.Geng Y, Aguilar‐Jakthong JS & Moatamed NA. Comparison of Afirma gene expression classifier with gene sequencing classifier in indeterminate thyroid nodules: a single‐institutional experience. Cytopathology 2020. 32 187–191. ( 10.1111/cyt.12920) [DOI] [PubMed] [Google Scholar]
  • 32.Polavarapu P, Fingeret A, Yuil-Valdes A, et al. Comparison of Afirma GEC and GSC to nodules without molecular testing in cytologically indeterminate thyroid nodules. J Endocr Soc 2021. 5 bvab148. ( 10.1210/jendso/bvab148) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Zhang L, Smola B, Lew M, et al. Performance of Afirma genomic sequencing classifier vs gene expression classifier in Bethesda category III thyroid nodules: an institutional experience. Diagn Cytopathol 2021. 49 921–927. ( 10.1002/dc.24765) [DOI] [PubMed] [Google Scholar]
  • 34.Gortakowski M, Feghali K & Osakwe I. Single institution experience with Afirma and Thyroseq testing in indeterminate thyroid nodules. Thyroid 2021. 31 1376–1382. ( 10.1089/thy.2020.0801) [DOI] [PubMed] [Google Scholar]
  • 35.Jin X, Lew M, Liron P, et al. Performance of Afirma genomic sequencing classifier and histopathological outcome in Bethesda category III thyroid nodules: initial versus repeat fine‐needle aspiration. Diagn Cytopathol 2023. 51 698–704. ( 10.1002/dc.25203) [DOI] [PubMed] [Google Scholar]
  • 36.Jin X, Lew M, Pantanowitz L, et al. Performance of Afirma genomic sequencing classifier and histopathological outcome are associated with patterns of atypia in Bethesda category III thyroid nodules. Cancer Cytopathology 2022. 130 891–898. ( 10.1002/cncy.22625) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Yang Z, Zhang T, Layfield L, et al. Performance of Afirma gene sequencing classifier versus gene expression classifier in thyroid nodules with indeterminate cytology. J Am Soc Cytopathology 2022. 13 458. ( 10.1016/j.jasc.2021.07.002) [DOI] [PubMed] [Google Scholar]
  • 38.Ahmadi S, Kotwal A, Bikas A, et al. Outcomes of cytologically indeterminate thyroid nodules managed with genomic sequencing classifier. J Clin Endocrinol Metab 2024. 109 e2231–e2239. ( 10.1210/clinem/dgae112) [DOI] [PubMed] [Google Scholar]
  • 39.Lončar I, van Velsen EFS, Massolt ET, et al. European experience with the Afirma gene expression classifier for indeterminate thyroid nodules: a clinical utility study in the Netherlands. Head Neck 2023. 45 2227–2236. ( 10.1002/hed.27472) [DOI] [PubMed] [Google Scholar]
  • 40.Jug R, Foo W, Jones C, et al. High‐risk and intermediate‐high–risk results from the ThyroSeq v2 and v3 thyroid genomic classifier are associated with neoplasia: independent performance assessment at an academic institution. Cancer Cytopathology 2020. 128 563–569. ( 10.1002/cncy.22283) [DOI] [PubMed] [Google Scholar]
  • 41.Carty SE, Paul Ohori N, Hilko DA, et al. The clinical utility of molecular testing in the management of thyroid follicular neoplasms (Bethesda IV nodules). Ann Surg 2020. 272 621–627. ( 10.1097/sla.0000000000004130) [DOI] [PubMed] [Google Scholar]
  • 42.Chen T, Gilfix BM, Rivera J, et al. The role of the ThyroSeq v3 molecular test in the surgical management of thyroid nodules in the Canadian public health care setting. Thyroid 2020. 30 1280–1287. ( 10.1089/thy.2019.0539) [DOI] [PubMed] [Google Scholar]
  • 43.Desai D, Lepe M, Baloch ZW, et al. ThyroSeq v3 for Bethesda III and IV: an institutional experience. Cancer Cytopathology 2020. 129 164–170. ( 10.1002/cncy.22362) [DOI] [PubMed] [Google Scholar]
  • 44.Gajzer DC, Tjendra Y, Kerr DA, et al. Probability of malignancy as determined by ThyroSeq v3 genomic classifier varies according to the subtype of atypia. Cancer Cytopathology 2022. 130 881–890. ( 10.1002/cncy.22617) [DOI] [PubMed] [Google Scholar]
  • 45.Youley T, Kerr DA, Zuo Y, et al. Probability of malignancy and molecular alterations as determined by ThyroSeq v3 genomic classifier in Bethesda category IV. Cancer Cytopathology 2023. 131 586–595. ( 10.1002/cncy.22737) [DOI] [PubMed] [Google Scholar]
  • 46.Sirotnikov S, Griffith CC, Lubin D, et al. ThyroSeq overview on indeterminate thyroid nodules: an institutional experience. Diagn Cytopathol 2024. 52 353–361. ( 10.1002/dc.25311) [DOI] [PubMed] [Google Scholar]
  • 47.Yang SP, Nga ME, Bundele MM, et al. Performance of a multigene genomic classifier and clinical parameters in predicting malignancy in a Southeast Asian cohort of patients with cytologically indeterminate thyroid nodules. Cancer Cytopathology 2024. 132 309–319. ( 10.1002/cncy.22796) [DOI] [PubMed] [Google Scholar]
  • 48.Lévesque F, Payne R, Beaudoin D, et al. Publicly funded molecular testing of indeterminate thyroid nodules: canada’s experience. J Clin Endocrinol Metab 2025. 110 e1031–e1037. ( 10.1210/clinem/dgae355) [DOI] [PubMed] [Google Scholar]
  • 49.Vardarli I, Tan S, Görges R, et al. Diagnostic accuracy of Afirma gene expression classifier, Afirma gene sequencing classifier, ThyroSeq v2 and ThyroSeq v3 for indeterminate (Bethesda III and IV) thyroid nodules: a meta-analysis. Endocr Connections 2024. 13 e240170. ( 10.1530/ec-24-0170) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.White MK, Thedinger WB & Dhingra JK. Long-term Follow-up of cytologically indeterminate thyroid nodules found benign on molecular testing: a validation study. OTO Open 2022. 6 2473974X2210835. ( 10.1177/2473974x221083542) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Schumm MA, Nguyen DT, Kim J, et al. Longitudinal assessment of quality of life following molecular testing for indeterminate thyroid nodules. Ann Surg Oncol 2021. 28 8872–8881. ( 10.1245/s10434-021-10375-6) [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplementary_materials.pdf (591.5KB, pdf)

Articles from European Thyroid Journal are provided here courtesy of Bioscientifica Ltd.

RESOURCES