Abstract
Objective
To determine the proportion of aspirates reclassified into each Bethesda category and to assess the rates of malignancy in each of them on repeat fine-needle aspiration biopsy (RFNA) following an AUS/FLUS diagnosis.
Design
Systematic review and meta-analysis
Methods
On February 2019, Pubmed/MEDLINE, EMBASE, WoS, and the Cochrane Library were searched for articles published from January 1, 2007. All studies published in English describing RFNA outcomes in AUS/FLUS nodules were included. PRISMA and MOOSE guidelines were followed. Five investigators independently assessed the eligibility of the studies. Two investigators extracted summary data and assessed the risk of bias. Data were pooled using a random-effects model. The rate of malignancy was calculated on resected nodules only (upper limit of true value); and considering all unresected nodules were benign (lower limit of true value). The protocol was registered in PROSPERO (CRD42019123114).
Results
Of 2937 retrieved studies, 27 were eligible. The meta-analysis was conducted on summary data of 3932 AUS/FLUS thyroid nodules with RFNA. RFNA cytology would reclassify into categories I through VI of Bethesda: 4% (3%, 5%), 48% (43%, 54%), 26% (20%, 32%), 4% (3%, 6%), 5% (3%, 6%), and 2% (1%, 2%) of AUS/FLUS nodules. Malignancy rates of resected nodules were 24% (9%, 38%), 4% (1%, 7%), 40% (28%, 52%), 37% (27%, 47%), 79% (69%, 90%), and 99% (95%, 100%) for categories I through VI of Bethesda. There was high heterogeneity in these data.
Conclusions
RFNA reclassified two-thirds of the AUS/FLUS specimens into a more definitive cytological category, with a benign call rate of nearly 50% and a negative predictive value greater than 96%.
Introduction
Fine needle aspiration (FNA) cytology has provided the best presurgical diagnostic approximation to thyroid nodules since the 1980s. A decade ago, the development of The Bethesda System for Reporting Thyroid Cytopathology (Bethesda) allowed thyroid cytology reporting and terminology standardization improving dramatically the clinical management of patients with thyroid nodules (1). Nevertheless, decision-making continued to be challenging for thyroid nodules with cytological diagnosis of atypia/follicular lesion of undetermined significance (AUS/FLUS).
The longstanding management recommendation for this cytology category was to repeat the FNA (RFNA) (1, 2). However, the 2015 American Thyroid Association guidelines for the management of adult patients with thyroid nodules and differentiated thyroid cancer (ATA guidelines) recognized molecular marker testing as an alternative to RFNA. This recommendation’s change was triggered by the suggestion of unreliable benign cytological results on RFNA following an initial AUS/FLUS diagnosis (3) and buttressed by molecular markers’ clinical validation studies promising results, which reached a negative predictive value (NPV) around 95% in AUS/FLUS (4, 5, 6, 7) If benign cytological diagnosis on RFNA is indeed unreliable, management of patients with AUS/FLUS thyroid nodules would be significantly impacted in many countries where molecular markers are not widely available yet.
To date, very few published studies have focused on reporting RFNA outcomes for AUS/FLUS aspirates. However, this information has been detailed in several manuscripts. This systematic review and meta-analysis aimed to retrieve and summarize this data to determine the proportion of aspirates reclassified into each Bethesda category and the rate of malignancy for each Bethesda category after RFNA following an initial AUS/FLUS cytology. In particular, we seek to establish the proportion of nodules reclassified as benign (benign call rate), and whether this result is reliable (negative predictive value greater than 95%).
Methods
This systematic review and meta-analysis followed the PRISMA and MOOSE guidelines and were not funded (8, 9). The protocol was registered in PROSPERO with the number CRD42019123114. The study was judged exempt from review by the Hospital Universitario Ramón y Cajal Ethics Committee.
Search strategy and selection criteria
PubMed/MEDLINE, Web of Science, the Cochrane Library, and EMBASE databases were systematically searched from January 1, 2007 up to the date of search in February 2019. The search strategy was developed by an investigator (P V) and a librarian (S C S) (Supplementary Table 1, see section on supplementary materials given at the end of this article). A combination of free-text-keywords, MeSH and database-specific controlled vocabulary was used to increase the sensitivity of the search in each database. Search results were limited to articles published after January 2007, which is the year in which the NCI FNA State of the Science Conference was held. Search results were de-duplicated twice, once using EndNote and once upon entry into Covidence, which was used throughout the study selection process.
Five reviewers (A B, P B, C A, V D T, and P V) screened the articles by title and abstract against the inclusion and exclusion criteria listed below. Each title and abstract were screened by two reviewers independently. Conflicting results were resolved by consensus or by a third-party author. The authors conducted a manual review of the references listed within articles selected for full-text inclusion.
We included all original studies in English with cytological and/or histological outcomes of AUS/FLUS nodules with repeat fine-needle aspiration biopsy in humans with no restrictions of age, gender, ethnicity, or any other health status information. Studies were excluded from the analysis if (i) they did not provide data on repeat biopsy, (ii) did not use Bethesda classification, or (iii) used core needle biopsy instead of FNA. We also excluded articles: (iv) in which specimens were not consecutively collected, (v) had overlapping cohort with a larger study, (vi) included nodules with known clinical or molecular evidence of malignancy, or (vii) in which data from RFNA of AUS/FLUS nodules could not be separated from RFNA of other categories.
Data extraction and risk of bias assessment
Summary data were extracted independently by two investigators (A B and P B). Disagreements were discussed and resolved by consensus after each step of the selection process. We extracted information for the following variables: study setting, study population and participant demographics and baseline characteristics, study methodology, study dates, information for the assessment of the risk of bias, results of initial biopsy (number and proportion of aspirates classified into each Bethesda category), results of repeat biopsy in AUS/FLUS nodules (number and proportion of aspirates classified into each Bethesda category), resection rates of each cytological category after single biopsy, resection rates of each cytological category on repeat biopsy, rates of malignancy (histologically proven) of each cytological category in nodules resected after the initial biopsy, rates of malignancy (histologically proven) of each cytological category on repeat biopsy, rates of nodules observed in each category after initial cytological diagnosis or after repeat biopsy following an AUS/FLUS diagnosis, time of follow-up (active surveillance) without evidence of malignancy. Corresponding authors of studies with missing information were contacted and given the opportunity to provide such information. Risk of bias was assessed by two investigators independently (A B and P B), using the Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies of the National Heart, Lung, and Blood Institute (10). All disagreements were resolved by consensus.
Demographic and design characteristics of studies included in the meta-analyses.
| Reference | Country | Design | Multi-centric | Study period | Study cohort | Study goal | AUS/FLUS (n, %a) | AUS/FLUS w/ RFNA (n) |
|---|---|---|---|---|---|---|---|---|
| (12) | Saudi Arabia | R | Yes | 2010–2014 | AUS/FLUS and FN | Evaluate outcomes in AUS/FLUS and FN | 115 (4.4) | 9 |
| (13)b | Saudi Arabia | R | No | 2011–2014 | AUS/FLUS | Identify AUS/FLUS clinical and US features associated with histological diagnosis | 305 (13.6) | 47 |
| (14) | USA | R | No | 2012–2014 | AUS | Evaluate AUS/FLUS management and outcomes | 976 (8.5) | 281 |
| (15) | USA | R | No | 2009–2012 | AUS/FLUS | Evaluate AUS/FLUS management and outcomes | 322 (13.9) | 101 |
| (16) | India | R | No | 2012–2016 | AUS/FLUS | Identify AUS/FLUS clinical and US features associated with histological diagnosis | 63 (6.5) | 25 |
| (17) | USA | R | No | 2010–2011 | A/FLUS | Assess the effect of TBSRTC on AUS/FLUS cytological diagnosis, management, and outcomes. | 61 (15.5) | 26 |
| (18)b | Spain | R | No | 2010–2014 | AUS/FLUS | Review institutional management of AUS/FLUS | 151 (5.0) | 61 |
| (19) | USA | R | Yes | 2005–2007 | AUS/FLUS and FN | Evaluate AUS/FLUS and FN management and outcomesc | 509 (NA) | 203 |
| (20)b | Singapore | R | No | 2008–2014 | AUS/FLUS | Assess malignancy rates of AUS/FLUS with or without nuclear atypia | 309 (6.4) | 73 |
| (21) | Turkey | R | No | 2010–2013 | AUS | Evaluate AUS/FLUS management and outcomes | 347 (7.1) | 118 |
| (22) | USA | R | No | 2008–2011 | AUS/FLUS | Assess malignancy rates of AUS/FLUS | 709 (8.0) | 96 |
| (23) | Korea | R | No | 2011–2014 | AUS/FLUS | Evaluate impact of clinical and US features on AUS/FLUS malignancy rates | 687 (10.8) | 248 |
| (24)b | Taiwan | R | No | 2012–2016 | AUS/FLUS | Describe distribution of thyroid FNA cytology and analyze AUS/FLUS outcomes | 909 (3.0) | 146 |
| (25) | Turkey | R | No | 2011–2015 | AUS/FLUS | Evaluate impact of RFNA on AUS/FLUS management and outcomes | 607 (11.8) | 171 |
| (26) | Republic of North Macedonia | R | No | 2012–2016 | AUS/FLUS | Evaluate impact of clinical and US features on AUS/FLUS malignancy rates | 281 (5.9) | 21 |
| (27) | Iran | P | No | 2013–2014 | AUS/FLUS | Evaluate the diagnostic value of RFNA in AUS/FLUS undergoing thyroidectomy | 50 (NA) | 50 |
| (28) | Korea | R | No | 2010–2013 | AUS/FLUS | Evaluate outcomes of nodules with two AUS/FLUS results | 441 (6.6) | 236 |
| (29) | Brazil | P | No | 2009–2013 | AUS/FLUS | Identify malignancy predictors in AUS/FLUS | 162 (9.9) | 150 |
| (30) | Brazil | R | No | 2011–2015 | B-I and AUS/FLUS | Evaluate the diagnostic value of RFNA on B-I and AUS/FLUS | 41 (23.0) | 41 |
| (31) | Australia | R | No | 2010–2013 | All thyroid FNAs | Determine malignancy rates of Bethesda categories | 97 (4.7) | 12 |
| (32) | USA | R | No | 2015 (Jan–Dec) | AUS/FLUS | Evaluate AUS/FLUS management and outcomes | 43 (4.8) | 12 |
| (33) | USA | R | No | 2006–2012 | AUS/FLUS | Review institutional AUS/FLUS malignancy rates | 221 (9.7) | 111 |
| (34) | Czech Republic | R | No | 1998–2012 | All thyroid FNAs | Assess the effect of TBSRTC on different cytological categories and their malignancy risk | 111 (8.5) | 102 |
| (35) | Turkey | R | Yes | 2011–2015 | AUS/FLUS | Identify malignancy predictors in AUS/FLUS | 1019 (10.3) | 518 |
| (36)b | Turkey | R | No | 2007–2011 | AUS/FLUS and FN | Review institutional outcomes in AUS/FLUS and FN | 1555 (15.9) | 544 |
| (3) | USA | R | No | 2005–2009 | AUS | Review RFNA experience and outcomes in AUS/FLUS | 512 (10.9) | 287 |
| (37) | Korea | R | No | 2010–2012 | AUS/FLUS | Identify malignancy predictors in AUS/FLUS | 772 (6.4) | 243 |
aPercent of AUS/FLUS nodules out of all thyroid nodules with cytological evaluation during the study period. bData in the last two columns of the table were provided by the study’s contact information author to include only nodules meeting inclusion criteria. cIn this study, nodules were originally classified using a 6-tier classification system similar to that of TBSRTC, and aspirates were reclassified into TBSRTC AUS/FLUS and FN categories based on the review of previous reports.
AUS/FLUS, atypia/follicular lesion of undetermined significance (Bethesda category III); B-I, non-diagnostic cytological category (Bethesda category I); FN, follicular neoplasm (Bethesda category IV); FNA, fine-needle aspiration biopsy; NA, not available; P, prospective design; R, retrospective design; RFNA, repeat FNA; SFM, suspicious for malignancy (Bethesda category V); TBSRTC, The Bethesda System for Reporting Thyroid Cytopathology; US, ultrasonographic; w/, with.
Data synthesis and analysis
All studies meeting the inclusion criteria were included in the analysis. Stata software (version 16.1) was used to analyze aggregated data, and prevalences were estimated using the random-effects model, package metaprop. The I2 statistic was used to evaluate the degree of heterogeneity of the studies. Publication bias for the prevalence of malignancy of nodules with benign cytology following an AUS/FLUS diagnosis was evaluated using the Egger’s test and funnel plots of study size against log odds as previously described for meta-analysis of proportions (11). To avoid excluding studies without events, 0.05 events were added to all studies for the publication bias analysis. Univariable mixed-effects models (with study-level moderators) were used to explain the observed heterogeneity in the prevalence of malignancy of nodules with benign cytology following an AUS/FLUS diagnosis. In particular, the following moderators were explored: proportion of nodules with AUS/FLUS diagnosis on initial biopsy, proportion of nodules with AUS/FLUS subject to repeat biopsy, and proportion of resected nodules. Estimating malignancy rates on resected nodules only might result in an overestimation as nodules with non-malignant cytological diagnosis but clinically suspicious features might be more likely operated on. To mitigate this possible bias, we also estimated malignancy rates for each cytological category considering all unresected nodules were benign. Although this likely results in underestimation, the true rate of malignancy probably lies in between these two scenarios that should be considered as the upper and lower limits of the true malignancy rate.
Results
Study characteristics and risk of bias assessment
Of 2937 identified citations through literature search after duplicates were removed, 2826 were deemed ineligible following title and abstract screening. Twenty-seven of 111 studies selected for full-text review met inclusion criteria for the meta-analysis (Fig. 1 and Supplementary Table 2 for exclusion criteria of the other 84). Of the 27 included studies (Table 1), 23 were conducted at Academic centers, 2 enrolled patients prospectively whereas the rest had a retrospective design, and 3 were multicentric. Most studies had a fair (n = 22, 81%) quality rating in the risk of bias assessment (Supplementary Table 3). Most studies were conducted in the US (n = 8), but there were also studies from countries in Australia (n = 1), Asia (n = 13), Europe (n = 3), and South America (n = 2). After the initial AUS/FLUS cytology, 41% of nodules (95% CI: 33%, 49%; I2 94.57%) were sent for surgical resection without repeating the FNA. The rate of malignancy among these nodules was 35% (95% CI: 26%, 44%; I2 96.92%). On the other hand, 42% of the nodules (95% CI: 35%, 50%; I2 98.41%) underwent repeat FNA.
Study selection process.
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
Study selection process.
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
Study selection process.
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
Total number of AUS/FLUS nodules included in the meta-analyses of each Bethesda category after RFNA.
| Reference | Second cytological diagnosis | Number resected | Number with malignant histology | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B-I | B-II | B-III | B-IV | B-V | B-VI | B-I | B-II | B-III | B-IV | B-V | B-VI | B-I | B-II | B-III | B-IV | B-V | B-VI | |
| (12) | 0 | 8 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 |
| (13)a | 0 | 11 | 20 | 4 | 7 | 5 | 0 | 11 | 20 | 4 | 7 | 5 | 0 | 3 | 10 | 2 | 7 | 5 |
| (14) | 10 | 159 | 81 | 24 | 2 | 5 | 1 | 4 | 32 | 15 | 2 | 3 | 1 | 0 | 14 | 4 | 2 | 3 |
| (15) | 0 | 43 | 49 | 4 | 4 | 1 | 0 | 5 | 39 | 4 | 4 | 1 | 0 | 0 | 6 | 1 | 2 | 1 |
| (16) | 0 | 17 | 3 | 0 | 0 | 5 | 0 | 0 | 3 | 0 | 0 | 5 | 0 | 0 | 3 | 0 | 0 | 5 |
| (17) | 3 | 11 | 6 | 2 | 4 | 0 | 2 | 0 | 5 | 2 | 4 | 0 | NA | NA | 1 | NA | NA | NA |
| (18)a | 11 | 28 | 8 | 13 | 1 | 0 | 0 | 6 | 3 | 8 | 1 | 0 | 0 | 0 | 0 | NA | NA | 0 |
| (19) | 2 | 125 | 46 | 20 | 7 | 3 | 2 | 24 | 32 | 20 | 7 | 3 | 1 | 0 | 7 | 8 | 5 | 3 |
| (20)a | 4 | 44 | 20 | 2 | 2 | 1 | 0 | 2 | 6 | 2 | 1 | 1 | 0 | 0 | 3 | 0 | 1 | 1 |
| (21) | 28 | 65 | 17 | 3 | 4 | 1 | 7 | 7 | 5 | 1 | 4 | 1 | 2 | 2 | 0 | 1 | 3 | 1 |
| (22) | 5 | 41 | 37 | 5 | 4 | 4 | 2 | 2 | 19 | 4 | 2 | 2 | 0 | 0 | 5 | 2 | 0 | 2 |
| (23) | 29 | 123 | 49 | 6 | 18 | 23 | 1 | 4 | 4 | 3 | 15 | 18 | 0 | 3 | 2 | 1 | 15 | 18 |
| (24)a | 18 | 22 | 47 | 4 | 29 | 26 | 11 | 12 | 32 | 4 | 28 | 26 | 5 | 2 | 14 | 2 | 24 | 25 |
| (25) | 35 | 63 | 41 | 9 | 21 | 2 | 6 | 4 | 23 | 5 | 17 | 2 | 1 | 0 | 6 | 2 | 11 | 2 |
| (26) | 0 | 16 | 0 | 1 | 4 | 0 | 0 | 0 | 0 | 1 | 4 | 0 | 0 | 0 | 0 | 0 | 4 | 0 |
| (27) | 0 | 15 | 22 | 0 | 0 | 13 | 0 | 15 | 22 | 0 | 0 | 13 | 0 | 8 | 15 | 0 | 0 | 13 |
| (28) | 23 | 111 | 58 | 3 | 23 | 18 | NA | NA | 21 | NA | NA | NA | NA | NA | 18 | NA | NA | NA |
| (29) | 2 | 54 | 73 | 11 | 10 | 0 | 2 | 39 | 73 | 11 | 10 | 0 | 0 | 2 | 21 | 3 | 8 | 0 |
| (30) | 2 | 27 | 10 | 0 | 2 | 0 | 2 | 5 | 6 | 0 | 1 | 6 | 1 | 0 | 5 | 0 | 1 | 6 |
| (31) | 1 | 8 | 1 | 2 | 0 | 0 | 0 | 0 | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 |
| (32) | 0 | 8 | 2 | 0 | 0 | 2 | 0 | 1 | 2 | 0 | 0 | 2 | 0 | 0 | 1 | 0 | 0 | 1 |
| (33) | 13 | 50 | 30 | 8 | 9 | 1 | 3 | 7 | 20 | 6 | 7 | 1 | 2 | 2 | 3 | 3 | 4 | 0 |
| (34) | 0 | 49 | 38 | NA | NA | 0 | 0 | 8 | 27 | NA | NA | 0 | 0 | 0 | 3 | NA | NA | 0 |
| (35) | 74 | 186 | 220 | 12 | 22 | 4 | 14 | 18 | 74 | 10 | 14 | 4 | 2 | 3 | 29 | 6 | 8 | 4 |
| (36)a | 33 | 293 | 109 | 67 | 18 | 24 | 11 | 61 | 73 | 62 | 18 | 24 | 2 | 0 | 29 | 23 | 16 | 23 |
| (3) | 12 | 139 | 80 | 25 | 26 | 5 | 2 | 7 | 51 | 22 | NA | NA | 0 | 2 | 22 | 9 | NA | NA |
| (37) | 0 | 104 | 67 | NA | NA | NA | 0 | 11 | 50 | NA | NA | NA | 0 | 6 | 38 | NA | NA | NA |
aData provided by the study’s contact information author to include only nodules meeting inclusion criteria.
NA, not available.
Distribution of cytological categories on RFNA following an AUS/FLUS diagnosis
Twenty seven studies comprising 3932 AUS/FLUS nodules with RFNA were used to calculate the expected prevalence of cytological categories on RFNA following an initial AUS/FLUS diagnosis (3, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37) (Table 2). Using a random-effects model, the expected redistribution of nodules among the Bethesda categories after RFNA was 4% (95% CI: 3%, 5%) as non-diagnostic (B-I); 48% (95% CI: 43%, 54%) as benign (B-II); 26% (95% CI: 20%, 32%) as AUS/FLUS (B-III); 4% (95% CI: 3%, 6%) as follicular neoplasm (B-IV); 5% (95% CI: 3%, 6%) as suspicious for malignancy (B-V); 2% (95% CI: 1%, 2%) as malignant (B-VI). There was high heterogeneity between studies (I2>80% for all categories) (Supplementary Table 4).
Malignancy rates of cytological categories on RFNA following an AUS/FLUS diagnosis
A total of 27 studies comprising 1411 nodules were used to analyze the malignancy rates of each cytological category on RFNA following an initial AUS/FLUS diagnosis (Table 2). The adjusted overall proportion of resected AUS/FLUS nodules with RFNA was 42% (95% CI: 29–54%), although there was great variability between studies (I2 > 99%), with resection rates ranging between 11% and 100% for individual studies. The adjusted resection rates for each cytological category were 35% (95% CI: 21%, 48%) for B-I; 20% (95% CI: 6%, 34%) for B-II; 63% (95% CI: 54%, 71%) for B-III; 92% (95% CI: 87%, 97%) for B-IV; 97% (95% CI: 94%, 100%) for B-V; 100% (95% CI: 99%, 100%) for B-VI nodules. Heterogeneity for resection rates was large for B-I, BII, and B-III categories (I2 >94%), moderate for B-IV and B-V (I2 66% and 45%, respectively), and absent for B-VI nodules (I2 0%) (Supplementary Table 5). Among resected nodules (upper limit of true malignancy rate), the adjusted rates of malignancy for each cytological category were 24% (95% CI: 9–38%; I2 65.9%) for B-I; 4% (95% CI: 1–7%; I2 68.0%) for B-II; 40% (95% CI:28–52%; I2 93.0%) for B-III; 37% (95% CI: 27–47%; I2 48.5%) for B-IV; 79% (95% CI: 69–90%; I2 87.9%) for B-V; 99% (95% CI: 95–102%; I2 41.3%) for B-VI nodules (Fig. 2 and Supplementary Table 6).
Adjusted upper and lower limits of malignancy rates with 95% CI for each Bethesda category on RFNA following an AUS/FLUS diagnosis. The upper limits of the malignancy rates were calculated on resected nodules only. The lower limits of the malignancy rates were calculated considering benign all unresected nodules in each category. The 'true malignancy rate' is expected to fall between the upper and lower limits. B-I, non-diagnostic; B-II, benign; B-III, atypia/follicular lesion of undetermined significance (AUS/FLUS); B-IV; follicular neoplasm; B-V, suspicious for malignancy; B-VI, malignant.
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
Adjusted upper and lower limits of malignancy rates with 95% CI for each Bethesda category on RFNA following an AUS/FLUS diagnosis. The upper limits of the malignancy rates were calculated on resected nodules only. The lower limits of the malignancy rates were calculated considering benign all unresected nodules in each category. The 'true malignancy rate' is expected to fall between the upper and lower limits. B-I, non-diagnostic; B-II, benign; B-III, atypia/follicular lesion of undetermined significance (AUS/FLUS); B-IV; follicular neoplasm; B-V, suspicious for malignancy; B-VI, malignant.
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
Adjusted upper and lower limits of malignancy rates with 95% CI for each Bethesda category on RFNA following an AUS/FLUS diagnosis. The upper limits of the malignancy rates were calculated on resected nodules only. The lower limits of the malignancy rates were calculated considering benign all unresected nodules in each category. The 'true malignancy rate' is expected to fall between the upper and lower limits. B-I, non-diagnostic; B-II, benign; B-III, atypia/follicular lesion of undetermined significance (AUS/FLUS); B-IV; follicular neoplasm; B-V, suspicious for malignancy; B-VI, malignant.
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
The adjusted rates of malignancy for each cytological category considering all unresected nodules benign (lower limit of true malignancy rate) were 2% (95% CI:0–4%; I2 24.3%) for B-I; 0% (95% CI: 0–0%; I2 45.4%) for B-II; 25% (95% CI: 17–32%; I2 94.3%) for B-III; 28% (95% CI: 21–35%; I2 20.1%) for B-IV; 68% (95% CI: 48–88%; I2 95.1%) for B-V; 94% (95% CI: 89–100%; I2 60.74%) for B-VI nodules (Fig. 2 and Supplementary Table 7).
The study explored sources of heterogeneity for our main outcome, the prevalence of malignancy of nodules with benign cytology following an AUS/FLUS diagnosis. We found no evidence of publication bias based on either Egger’s test (P = 0.77) or the funnel plot (Fig. 3). Differences between studies could not be explained by either: (i) proportion of AUS/FLUS on initial biopsy (P = 0.60), (ii) proportion of AUS/FLUS with a repeat biopsy (P = 0.48) or (iii) rate of resection (P = 0.89).
Funnel plot for the prevalence of malignancy on resected nodules with benign cytology following an AUS/FLUS diagnosis (study size against log odds).
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
Funnel plot for the prevalence of malignancy on resected nodules with benign cytology following an AUS/FLUS diagnosis (study size against log odds).
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
Funnel plot for the prevalence of malignancy on resected nodules with benign cytology following an AUS/FLUS diagnosis (study size against log odds).
Citation: European Journal of Endocrinology 185, 4; 10.1530/EJE-21-0330
Discussion
This meta-analysis found that RFNA in nodules with an initial AUS/FLUS diagnosis can reclassify almost two-thirds of the nodules into another category. In particular, around 50% of the nodules would be reclassified as benign, and the rest into a higher-suspicion category. Furthermore, the malignancy rate of each Bethesda category on RFNA was no different from the expected values for each category (38, 39). This finding suggests the diagnosis of the second cytology is reliable. In particular, the malignancy rate of nodules reclassified as benign on repeat biopsy lies between 0 and 4%. Thus, the negative predictive value of RFNA is greater than 96%, which is considered enough to offer observation in lieu of diagnostic surgery (2, 40).
In the last decade, several molecular marker tests have been developed and commercialized for the evaluation of thyroid nodules with indeterminate cytology. The use of such tests was rapidly adopted by clinicians in the United States; and has already changed the practice by significantly reducing the number of resected nodules with indeterminate cytology (41). Furthermore, the use of molecular markers has been supported by the American Thyroid Association, followed by many clinicians around the globe, as an alternative to RFNA. However, the diagnostic performance of the until recently gold-standard approach to AUS/FLUS nodules –RFNA – had been seldom studied. In fact, although molecular tests claim high diagnostic performance, it was unclear what the advantage over RFNA was. Our meta-analysis provides evidence about the diagnostic performance of the traditional approach which will allow the development of more accurate models of cost-effectiveness.
The diagnostic performance of RFNA seems similar if not superior to that of Afirma GEC, which has been for many years the leading test in the market (4). Most recent versions of molecular tests such as Afirma GSC and ThyroSeq v3 claim higher benign call rates than we found with the RFNA for AUS/FLUS (6, 7). However, predictive values of molecular tests in real-world seem to be lower than anticipated (42, 43, 44, 45). Here we provide real-world data on 3932 AUS/FLUS nodules with RFNA from 27 studies, showing that diagnostic surgery can be safely (NPV greater than 96%) avoided in nearly 50% of AUS/FLUS with benign cytology on RFNA. A similar rate (51%) of avoided surgeries was found when Afirma GSC and Thyroseq v3 were used to characterized cytologically indeterminate thyroid nodules in a recent clinical trial (46). Furthermore, RFNA allows to risk-stratify another 15% of AUS/FLUS that are reclassified into a higher-suspicion category. This rate is similar, if not superior, to that of current oncogene panels as the identification of higher-risk mutations is very rare in thyroid nodules with indeterminate cytology (44, 47) and clearly superior to that of gene classifiers that currently do not risk-stratify suspicious specimens.
RFNA seems a useful and reliable approach for AUS/FLUS; however, we identified great heterogeneity between studies which could not be explained by the proportion of nodules initially classified as AUS/FLUS, the proportion of AUS/FLUS with RFNA, or the rates of resection. We hypothesize that this is probably due to three factors:
-
The heterogeneity of the category. By definition, there are many cytological scenarios that can be classified into the AUS/FLUS category, such as cellular atypia, architectural atypia, oncocytic changes, and other types of atypia (48). It has been shown that different cytological scenarios are associated with different rates of malignancy, with different histological types of cancer and different diagnostic performances of the molecular test (42, 49, 50, 51). Thus, it is conceivable that different cytological scenarios also have different outcomes on RFNA.
-
Differences in cytology interpretation. Thyroid cytology is known to have substantial intra- and inter-observer variability (52). This is particularly true for the AUS/FLUS category, which technically limits with most other diagnostic categories (53). Because the probability of classifying a cytological specimen as AUS/FLUS is determined by the training and experience of the pathologist, AUS/FLUS cohorts of different institutions are likely comprised of a different mixture of cytological scenarios. Such differences should have an impact on outcomes.
-
Differences in histology interpretation. It has been known for years that there is a poor inter-observer agreement for the diagnosis of follicular pattern lesions (52, 54, 55). Several studies have shown that if non-invasive follicular thyroid neoplasms with papillary-like nuclear features (NIFTPs) were not considered malignant, the prevalence of malignancy of resected thyroid nodules with benign cytology would drop by at least half (56, 57). It has also been suggested that the diagnostic threshold for NIFTP varies widely among centers (44, 58). Many studies were published before the recognition of NIFTP as a new entity (59). Thus, we were unable to do a separate analysis excluding non-invasive tumors.
We were unable to explain the observed heterogeneity which could be due to several factors. Thus, it seems necessary that each center conducts an internal quality assessment of the outcomes of RFNA. Furthermore, even though benign cytology following an AUS/FLUS diagnosis seems reliable, other clinical considerations such as nodule’s growth rate, sonographic pattern, or presence of compressive symptoms should be considered when deciding the most appropriate management. Among cytologically indeterminate thyroid nodules, the clinical scenario and the sonographic pattern have been shown to allow risk-stratification (50, 60). Thus, these features should be considered when deciding patient management for indeterminate nodules (AUS/FLUS or follicular neoplasms) on RFNA too.
Limitations
Only manuscripts published in English were included in the analysis. Thus, relevant articles published in other languages were left out. The meta-analysis was conducted on study-level data, which limit the quality of the results, but there was no evidence of publication bias. We did not describe the demographic characteristics of the meta-analyses’ patients. This data were reported inconsistently across studies and was, mostly, not specific for AUS/FLUS nodules with RFNA. However, most studies reported a female sex predominance, around 80%, in their cohorts; a mean age around 50 years and a mean nodule size around 2 cm, ranging between 1.2 and 2.8 cm. Our study summarized the outcomes of RFNA in a large cohort of AUS/FLUS nodules. However, most studies were retrospective in which describing the outcomes of RFNA in AUS/FLUS was not the main goal. In fact, as part of the standard of care, only a fraction of all patients with AUS/FLUS underwent RFNA. Therefore, our results could suffer from selection bias. To minimize this bias, we only included cohorts of consecutively diagnosed AUS/FLUS nodules. Nonetheless, it is possible that nodules with higher clinical suspicion for malignancy were resected immediately after the initial cytological diagnosis and that less suspicious nodules underwent observation without RFNA. The rate of malignancy among AUS/FLUS nodules resected without RFNA, however, was similar to those resected after a repeated AUS/FLUS diagnosis and to that found in previous metanalyses (38, 39).
Moreover, not all nodules subject to RFNA were resected. In fact, the rates of resection were rather low (20%) among nodules with benign cytology on RFNA. To mitigate this possible selection bias, we estimated the upper and lower limit rate of malignancy, acknowledging that the true rate of malignancy must lie between the two. Furthermore, because of this unaddressed selection bias, our results might better reflect real-world outcomes of RFNA, as the same selection bias is expected to continue driving the management of AUS/FLUS.
Conclusions
In conclusion, a RFNA following an AUS/FLUS cytological diagnosis is an efficient and reliable way to reclassify up to two-thirds of the nodules into a more definitive category. In fact, 50% of the nodules are expected to be reclassified as benign with a negative predictive value greater than 96%. Thus, RFNA should continue to be considered the preferred management recommendation for the evaluation of AUS/FLUS nodules, particularly in countries in which molecular tests are not readily available. In countries in which molecular tests are available, cost-effectiveness studies should be compared to the cost and diagnostic performance of RFNA.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EJE-21-0330.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this study.
Funding
This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.
Author contribution statement
Study concept and design: Pablo Valderrabano, Bryan McIver. Acquisition, analysis, or interpretation of data: All authors. Statistical analysis: Alfonso Muriel. Drafting of the manuscript: Ane Bayona, Patricia Benavent, and Pablo Valderrabano. Critical revision of the manuscript for important intellectual content: All authors. All authors approved the final version of the manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Acknowledgement
The authors thank all the authors who shared additional information form their published cohorts necessary to conduct the current study.
References
- 1↑
Cibas ES, Ali SZ. The Bethesda system for reporting thyroid cytopathology. Thyroid 2009 19 1159–1165. (https://doi.org/10.1089/thy.2009.0274)
- 2↑
Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, Pacini F, Randolph GW, Sawka AM & Schlumberger M et al.2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid 2016 26 1–133. (https://doi.org/10.1089/thy.2015.0020)
- 3↑
VanderLaan PA, Marqusee E, Krane JF. Clinical outcome for atypia of undetermined significance in thyroid fine-needle aspirations: should repeated FNA be the preferred initial approach? American Journal of Clinical Pathology 2011 135 770–775. (https://doi.org/10.1309/AJCP4P2GCCDNHFMY)
- 4↑
Alexander EK, Kennedy GC, Baloch ZW, Cibas ES, Chudova D, Diggans J, Friedman L, Kloos RT, LiVolsi VA & Mandel SJ et al.Preoperative diagnosis of benign thyroid nodules with indeterminate cytology. New England Journal of Medicine 2012 367 705–715. (https://doi.org/10.1056/NEJMoa1203208)
- 5↑
Nikiforov YE, Carty SE, Chiosea SI, Coyne C, Duvvuri U, Ferris RL, Gooding WE, LeBeau SO, Ohori NP & Seethala RR et al.Impact of the multi-gene ThyroSeq next-generation sequencing assay on cancer diagnosis in thyroid nodules with atypia of undetermined significance/follicular lesion of undetermined significance cytology. Thyroid 2015 25 1217–1223. (https://doi.org/10.1089/thy.2015.0305)
- 6↑
Patel KN, Angell TE, Babiarz J, Barth NM, Blevins T, Duh QY, Ghossein RA, Harrell RM, Huang J & Kennedy GC et al.Performance of a genomic sequencing classifier for the preoperative diagnosis of cytologically indeterminate thyroid nodules. JAMA Surgery 2018 153 817–824. (https://doi.org/10.1001/jamasurg.2018.1153)
- 7↑
Steward DL, Carty SE, Sippel RS, Yang SP, Sosa JA, Sipos JA, Figge JJ, Mandel S, Haugen BR & Burman KD et al.Performance of a multigene genomic classifier in thyroid nodules with indeterminate cytology: a prospective blinded multicenter study. JAMA Oncology 2019 5 204–212 (https://doi.org/10.1001/jamaoncol.2018.4616).
- 8↑
Moher D, Liberati A, Tetzlaff J, Altman DG & The PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Medicine 2009 6 e1000097. (https://doi.org/10.1371/journal.pmed.1000097)
- 9↑
Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D, Becker BJ, Sipe TA, Thacker SB. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000 283 2008–2012. (https://doi.org/10.1001/jama.283.15.2008)
- 10↑
Anon. Study quality assessment tools | NHLBI, NIH (online). n.d. (available at: https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools). Accessed 1 December 2020.
- 11↑
Hunter JP, Saratzis A, Sutton AJ, Boucher RH, Sayers RD, Bown MJ. In meta-analyses of proportion studies, funnel plots were found to be an inaccurate method of assessing publication bias. Journal of Clinical Epidemiology 2014 67 897–903. (https://doi.org/10.1016/j.jclinepi.2014.03.003)
- 12↑
Al-Abbadi MA, Shareef SQ, Yousef MM, Almasri NM, Mustafa HE, Aljawad H, Ali JA, Groves A, Alsaihati Y. A follow-up study on thyroid aspirates reported as atypia of undetermined significance/follicular lesion of undetermined significance and follicular neoplasm/suspicious for follicular neoplasm: a multicenter study from the Arabian Gulf region. Diagnostic Cytopathology 2017 45 983–988. (https://doi.org/10.1002/dc.23805)
- 13↑
Alqahtani SA, Alsobhi S, Alsalloum RI, Najjar SN, Al-Hindi HN. Surgical outcome of thyroid nodules with atypia of undetermined significance and follicular lesion of undetermined significance in fine needle aspiration biopsy. World Journal of Endocrine Surgery 2017 9 100–103. (https://doi.org/10.5005/jp-journals-10002-1220)
- 14↑
Brandler TC, Aziz MS, Coutsouvelis C, Rosen L, Rafael OC, Souza F, Jelloul FZ, Klein MA. Young investigator challenge: atypia of undetermined significance in thyroid FNA: standardized terminology without standardized management-a closer look at repeat FNA and quality measures: AUS: repeat FNA and quality monitoring. Cancer Cytopathology 2016 124 37–43. (https://doi.org/10.1002/cncy.21648)
- 15↑
Broome JT, Cate F, Solorzano CC. Utilization and impact of repeat biopsy for follicular lesion/atypia of undetermined significance. World Journal of Surgery 2014 38 628–633. (https://doi.org/10.1007/s00268-013-2330-0)
- 16↑
Chandra S, Chandra H, Bisht SS. Malignancy rate in thyroid nodules categorized as atypia of undetermined significance or follicular lesion of undetermined significance: an institutional experience. Journal of Cytology 2017 34 144–148. (https://doi.org/10.4103/JOC.JOC_234_16)
- 17↑
Chen JC, Pace SC, Chen BA, Khiyami A, McHenry CR. Yield of repeat fine-needle aspiration biopsy and rate of malignancy in patients with atypia or follicular lesion of undetermined significance: the impact of the Bethesda system for reporting thyroid cytopathology. Surgery 2012 152 1037–1044. (https://doi.org/10.1016/j.surg.2012.08.052)
- 18↑
Estrada Muñoz L, Díaz del Arco C, Ortega Medina L, Fernández Aceñero MJ. Thyroid atypia/follicular lesion of undetermined significance: attitudes towards the diagnosis of Bethesda system III nodules. Acta Cytologica 2017 61 21–26. (https://doi.org/10.1159/000452156)
- 19↑
Faquin WC, Baloch ZW. Fine-needle aspiration of follicular patterned lesions of the thyroid: diagnosis, management, and follow-up according to National Cancer Institute (NCI) recommendations. Diagnostic Cytopathology 2010 38 731–739. (https://doi.org/10.1002/dc.21292)
- 20↑
Gan TRX, Nga ME, Lum JHY, Wong WM, Tan WB, Parameswaran R, Ngiam KY. Thyroid cytology-nuclear versus architectural atypia within the ‘atypia of undetermined significance/follicular lesion of undetermined significance’ Bethesda category have significantly different rates of malignancy: subclassifying ‘atypical’ thyroid FNACs. Cancer Cytopathology 2017 125 245–256. (https://doi.org/10.1002/cncy.21823)
- 21↑
Gocun PU, Karakus E, Bulutay P, Akturk M, Akin M, Poyraz A. What is the malignancy risk for atypia of undetermined significance? Three years’ experience at a university hospital in Turkey. Cancer Cytopathology 2014 122 604–610. (https://doi.org/10.1002/cncy.21434)
- 22↑
Ho AS, Sarti EE, Jain KS, Wang H, Nixon IJ, Shaha AR, Shah JP, Krus DH, Ghossein R & Fish SA et al.Malignancy rate in thyroid nodules classified as Bethesda category III (AUS/FLUS). Thyroid 2014 24 832–839. (https://doi.org/10.1089/thy.2013.0317)
- 23↑
Hong SH, Lee H, Cho MS, Lee JE, Sung YA, Hong YS. Malignancy risk and related factors of atypia of undetermined significance/follicular lesion of undetermined significance in thyroid fine needle aspiration. International Journal of Endocrinology 2018 2018 4521984. (https://doi.org/10.1155/2018/4521984)
- 24↑
Jan IS, Lee YT, Wang CM, Cheng TY, Wang CY, Chang TC, Shih SR. The surgery and repeat aspiration outcomes of the atypia of undetermined significance/follicular lesion of undetermined significance category in the Bethesda system for reporting thyroid cytopathology. Asian Journal of Surgery 2019 42 144–147. (https://doi.org/10.1016/j.asjsur.2018.02.008)
- 25↑
Kuru B, Atmaca A, Kefeli M. Malignancy rate associated with Bethesda category III (AUS/FLUS) with and without repeat fine needle aspiration biopsy. Diagnostic Cytopathology 2016 44 394–398. (https://doi.org/10.1002/dc.23456)
- 26↑
Mileva M, Stoilovska B, Jovanovska A, Ugrinska A, Petrushevska G, Kostadinova-Kunovska S, Miladinova D, Majstorov V. Thyroid cancer detection rate and associated risk factors in patients with thyroid nodules classified as Bethesda category III. Radiology and Oncology 2018 52 370–376. (https://doi.org/10.2478/raon-2018-0039)
- 27↑
Nazem M, Kaleydari B, Kolahdouzan M, Kashi V, Amini M. Fine-needle aspirations in atypia of unknown significance/follicular lesion undetermined significance (AUS/FLUS): usefulness of repeated FNA of thyroid nodules. International Journal of Cancer Management 2016 9 e7879. (https://doi.org/10.17795/ijcp-7879)
- 28↑
Park VY, Kim EK, Kwak JY, Yoon JH, Moon HJ. Malignancy risk and characteristics of thyroid nodules with two consecutive results of atypia of undetermined significance or follicular lesion of undetermined significance on cytology. European Radiology 2015 25 2601–2607. (https://doi.org/10.1007/s00330-015-3668-5)
- 29↑
Rosario PW Thyroid nodules with atypia or follicular lesions of undetermined significance (Bethesda category III): importance of ultrasonography and cytological subcategory. Thyroid 2014 24 1115–1120. (https://doi.org/10.1089/thy.2013.0650)
- 30↑
Saieg MA, Barbosa B, Nishi J, Ferrari A, Costa F. The impact of repeat FNA in non-diagnostic and indeterminate thyroid nodules: a 5-year single-centre experience. Cytopathology 2018 29 196–200. (https://doi.org/10.1111/cyt.12508)
- 31↑
Sarkis LM, Norlen O, Aniss A, Watson N, Delbrigde LW, Sidhu SB, Sywak MS, Gill AJ. The Australian experience with the Bethesda classification system for thyroid fine needle aspiration biopsies. Pathology 2014 46 592–595. (https://doi.org/10.1097/PAT.0000000000000168)
- 32↑
Seagrove‐Guffey MA, Hatic H, Peng H, Bates KC, Odugbesan AO. Malignancy rate of atypia of undetermined significance/follicular lesion of undetermined significance in thyroid nodules undergoing FNA in a suburban endocrinology practice: a retrospective cohort analysis. Cancer Cytopathology 2018 126 881–888. (https://doi.org/10.1002/cncy.22054)
- 33↑
Shrestha RT, Hennessey JV. Cytologic subclassification of atypia of undetermined significance may predict thyroid nodules more likely to be malignant at surgery. Diagnostic Cytopathology 2016 44 492–498. (https://doi.org/10.1002/dc.23472)
- 34↑
Srbova L, Gabalec F, Ryska A, Cap J. Results of retrospective classification of thyroid FNAs according to the Bethesda system: would this have improved accuracy? Cytopathology 2015 26 231–237. (https://doi.org/10.1111/cyt.12171)
- 35↑
Turkyilmaz S, Ulusahin M, Celebi B, Cekic AB, Mungan S, Kucuktulu U, Tasdelen A, Guner A, Cinel A. Thyroid nodules classified as atypia or follicular lesions of undetermined significance deserve further research: analysis of 305 surgically confirmed nodules. Cytopathology 2017 28 391–399. (https://doi.org/10.1111/cyt.12438)
- 36↑
Ustün H, Astarci HM, Altunkaya CY, Yilmaz S, Barin A, Ekici S, Caydere M. Fine-needle aspiration of follicular patterned lesions of the thyroid: diagnosis, management, and follow-up according to thyroid Bethesda system. Acta Cytologica 2012 56 361–369. (https://doi.org/10.1159/000338218)
- 37↑
Yoo MR, Gweon HM, Park AY, Cho KE, Kim JA, Youk JH, Son EJ. Repeat diagnoses of Bethesda category III thyroid nodules: what to do next? PLoS ONE 2015 10 e0130138. (https://doi.org/10.1371/journal.pone.0130138)
- 38↑
Bongiovanni M, Spitale A, Faquin WC, Mazzucchelli L, Baloch ZW. The Bethesda system for reporting thyroid cytopathology: a meta-analysis. Acta Cytologica 2012 56 333–339. (https://doi.org/10.1159/000339959)
- 39↑
Straccia P, Rossi ED, Bizzarro T, Brunelli C, Cianfrini F, Damiani D, Fadda G. A meta-analytic review of the Bethesda system for reporting thyroid cytopathology: has the rate of malignancy in indeterminate lesions been underestimated?: TBSRTC. Cancer Cytopathology 2015 123 713–722. (https://doi.org/10.1002/cncy.21605)
- 40↑
Ferris RL, Baloch Z, Bernet V, Chen A, Fahey TJ, Ganly I, Hodak SP, Kebebew E, Patel KN & Shaha A 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. (https://doi.org/10.1089/thy.2014.0502)
- 41↑
Duick DS, Klopper JP, Diggans JC, Friedman L, Kennedy GC, Lanman RB, MacIver B. The impact of benign gene expression classifier test results on the endocrinologist–patient decision to operate on patients with thyroid nodules with indeterminate fine-needle aspiration cytopathology. Thyroid 2012 22 996–1001. (https://doi.org/10.1089/thy.2012.0180)
- 42↑
Valderrabano P, Hallanger-Johnson JE, Thapa R, Wang X, McIver B. Comparison of postmarketing findings vs the initial clinical validation findings of a thyroid nodule gene expression classifier: a systematic review and meta-analysis. JAMA Otolaryngology: Head and Neck Surgery 2019 145 783–792. (https://doi.org/10.1001/jamaoto.2019.1449)
- 43↑
Valderrabano P, Khazai L, Leon ME, Thompson ZJ, Ma Z, Chung CH, Hallanger-Johnson JE, Otto KJ, Rogers KD & Centeno BA et al.Evaluation of ThyroSeq v2 performance in thyroid nodules with indeterminate cytology. Endocrine-Related Cancer 2017 24 127–136. (https://doi.org/10.1530/ERC-16-0512)
- 44↑
Marcadis AR, Valderrabano P, Ho AS, Tepe J, Swarztwelder CE, Byrd S, Sacks WL, Untch BR, Shaha AR & Xu B et al.Interinstitutional variation in predictive value of the ThyroSeq v2 genomic classifier for cytologically indeterminate thyroid nodules. Surgery 2019 165 17–24. (https://doi.org/10.1016/j.surg.2018.04.062)
- 45↑
Marti JL, Shaha AR. Molecular testing for indeterminate thyroid nodules-when the rubber meets the road. JAMA Otolaryngology: Head and Neck Surgery 2019 145 792–793. (https://doi.org/10.1001/jamaoto.2019.1465)
- 46↑
Livhits MJ, Zhu CY, Kuo EJ, Nguyen DT, Kim J, Tseng CH, Leung AM, Rao J, Levin M & Douek ML et al.Effectiveness of molecular testing techniques for diagnosis of indeterminate thyroid nodules: a randomized clinical trial. JAMA Oncology 2021 7 70–77. (https://doi.org/10.1001/jamaoncol.2020.5935)
- 47↑
Valderrabano P, Khazai L, Thompson ZJ, Leon ME, Otto KJ, Hallanger-Johnson JE, Wadsworth JT, Chung CH, Centeno BA, McIver B. Impact of oncogene panel results on surgical management of cytologically indeterminate thyroid nodules. Head and Neck 2018 40 1812–1823. (https://doi.org/10.1002/hed.25165)
- 48↑
Cibas ES, Ali SZ. The 2017 Bethesda system for reporting thyroid cytopathology. Journal of the American Society of Cytopathology 2017 6 217–222. (https://doi.org/10.1016/j.jasc.2017.09.002)
- 49↑
Valderrabano P, Khazai L, Thompson ZJ, Sharpe SC, Tarasova VD, Otto KJ, Hallanger-Johnson JE, Wadsworth JT, Wenig BM & Chung CH et al.Cancer risk associated with nuclear atypia in cytologically indeterminate thyroid nodules: a systematic review and meta-analysis. Thyroid 2018 28 210–219. (https://doi.org/10.1089/thy.2017.0419)
- 50↑
Valderrabano P, Khazai L, Thompson ZJ, Leon ME, Otto KJ, Hallanger-Johnson JE, Wadsworth JT, Wenig BM, Chung CH & Centeno BA et al.Cancer risk stratification of indeterminate thyroid nodules: a cytological approach. Thyroid 2017 27 1277–1284. (https://doi.org/10.1089/thy.2017.0221)
- 51↑
Brauner E, Holmes BJ, Krane JF, Nishino M, Zurakowski D, Hennessey JV, Faquin WC, Parangi S. Performance of the afirma gene expression classifier in Hürthle cell thyroid nodules differs from other indeterminate thyroid nodules. Thyroid 2015 25 789–796. (https://doi.org/10.1089/thy.2015.0049)
- 52↑
Cibas ES, Baloch ZW, Fellegara G, LiVolsi VA, Raab SS, Rosai J, Diggans J, Friedman L, Kennedy GC & Kloos RT et al.A prospective assessment defining the limitations of thyroid nodule pathologic evaluation. Annals of Internal Medicine 2013 159 325–332. (https://doi.org/10.7326/0003-4819-159-5-201309030-00006)
- 53↑
Ohori NP, Schoedel KE. Variability in the atypia of undetermined significance/follicular lesion of undetermined significance diagnosis in the Bethesda system for reporting thyroid cytopathology: sources and recommendations. Acta Cytologica 2011 55 492–498. (https://doi.org/10.1159/000334218)
- 54↑
Lloyd RV, Erickson LA, Casey MB, Lam KY, Lohse CM, Asa SL, Chan JK, DeLellis RA, Harach HR & Kakudo K et al.Observer variation in the diagnosis of follicular variant of papillary thyroid carcinoma. American Journal of Surgical Pathology 2004 28 1336–1340. (https://doi.org/10.1097/01.pas.0000135519.34847.f6)
- 55↑
Hirokawa M, Carney JA, Goellner JR, DeLellis RA, Heffness CS, Katoh R, Tsujimoto M, Kakudo K. Observer variation of encapsulated follicular lesions of the thyroid gland. American Journal of Surgical Pathology 2002 26 1508–1514. (https://doi.org/10.1097/00000478-200211000-00014)
- 56↑
Strickland KC, Howitt BE, Marqusee EA, Alexander EK, Cibas ES, Krane JF, Barletta JA. The impact of noninvasive follicular variant of papillary thyroid carcinoma on rates of malignancy for fine-needle aspiration diagnostic categories. Thyroid 2015 25 987–992. (https://doi.org/10.1089/thy.2014.0612)
- 57↑
Faquin WC, Wong LQ, Afrogheh AH, Ali SZ, Bishop JA, Bongiovanni M, Pusztaszeri MP, VandenBussche CJ, Gourmaud J & Vaickus LJ et al.Impact of reclassifying noninvasive follicular variant of papillary thyroid carcinoma on the risk of malignancy in the Bethesda system for reporting thyroid cytopathology. Cancer Cytopathology 2016 124 181–187. (https://doi.org/10.1002/cncy.21631)
- 58↑
Bychkov A, Hirokawa M, Jung CK, Liu Z, Zhu Y, Hong SW, Satoh S, Lai CR, Huynh L, Kakudo K. Low rate of noninvasive follicular thyroid neoplasm with papillary-like nuclear features in Asian practice. Thyroid 2017 27 983–984. (https://doi.org/10.1089/thy.2017.0079)
- 59↑
Nikiforov YE, Seethala RR, Tallini G, Baloch ZW, Basolo F, Thompson LDR, Barletta JA, Wenig BM, Ghuzlan AA & Kakudo K et al.Nomenclature revision for encapsulated follicular variant of papillary thyroid carcinoma: a paradigm shift to reduce overtreatment of indolent tumors. JAMA Oncology 2016 2 1023–1029. (https://doi.org/10.1001/jamaoncol.2016.0386)
- 60↑
Valderrabano P, McGettigan MJ, Lam CA, Khazai L, Thompson ZJ, Chung CH, Centeno BA, McIver B. Thyroid nodules with indeterminate cytology: utility of the American Thyroid Association sonographic patterns for cancer risk stratification. Thyroid 2018 28 1004–1012. (https://doi.org/10.1089/thy.2018.0085)
