Subclassification of the Bethesda Category III (AUS/FLUS): A study of thyroid FNA cytology based on ThinPrep slides from the National Cancer Center in China

Abstract

Background The atypia of an undetermined significance/follicular lesion of undetermined significance (AUS/FLUS) category in the Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) is a heterogeneous category, which includes various cell patterns. The subclassification of AUS/FLUS was suggested in the 2017 TBSRTC. However, the risk of malignancy (ROM) associated with different subgroups remains unresolved. Herein, AUS/FLUS aspirates were subclassified, from which the ROM of each subgroup was determined. Methods All cases undergoing fine‐needle aspiration (FNA) from July 2013 to December 2018 were reviewed. Of 12,913 thyroid FNAs, 1053 (8.2%) were AUS/FLUS. The slides of 222 patients with AUS/FLUS with surgical follow‐up were reviewed and subclassified according to the recommendations of the 2017 TBSRTC. There were 195 aspirates consistently diagnosed as AUS/FLUS and subclassified as cytologic atypia 1 (AUS‐C1); cytologic atypia 2 (AUS‐C2); architectural atypia (AUS‐A); cytologic and architectural atypia (AUS‐C&A); Hürthle cell aspirates (AUS‐H); atypia, not otherwise specified (AUS‐NOS); and atypical lymphoid cells, rule out lymphoma (AUS‐L). Results Malignancy was identified in 83.3% (185 of 222) of the AUS/FLUS nodules. The AUS‐C1 group was the most common (62.1%), followed by the AUS‐C&A (12.8%), AUS‐C2 (10.8%), AUS‐H (6.7%), AUS‐NOS (5.6%), AUS‐L (1.5%), and AUS‐A (0.5%) groups. AUS‐C1 had the highest ROM (92.6%) among the groups and varied significantly from that of the AUS‐C&A (P = .171), AUS‐C2 (P = .001), AUS‐H (P = .001), and AUS‐NOS (P < .001) groups. Conclusions As a heterogeneous category of TBSRTC, the ROM for AUS/FLUS varies greatly among medical centers. Subclassification of AUS/FLUS might be helpful in identifying nodules with a high ROM in this category and improving the management of such nodules.

Predictors of Malignancy in Thyroid Nodules Classified as Bethesda Category III

“…The malignancy rate in AUS/FLUS nodules varies from 17% to 83.3%, according to the current evidence ( 10 , 14 ). Although the rate in the present study was also within this range, it was surprising to note there is great variability in the incidence.…”

“…Although the rate in the present study was also within this range, it was surprising to note there is great variability in the incidence. In a study by Zhao et al (10), the authors attributed their high incidence rate to selection bias and strict cytological criteria for malignancy diagnosis. In our cancer center, rather than being based on molecular findings, a decision of surgical treatment was usually made if the clinical/radiological examination revealed malignancy.…”

“…More importantly, there is substantial evidence that the real malignant risk of AUS/FLUS is significantly higher than the estimated incidence (7)(8)(9)(10)(11)(12)(13). Therefore, it is important to identify reliable indicators, such as the BRAF V600E mutation status, adverse sonographic features, nuclear and architectural atypia, and circulating cell-free DNA integrity, for determining the likelihood of malignancy and guiding clinical decision-making (7)(8)(9)(10)(11)(12)(13),. However, to our knowledge, there is still controversy regarding the predictors of malignancy in Bethesda III group thyroid nodules.…”

Two‐tier subclassification of the Bethesda category III (atypia of undetermined significance/follicular lesion of undetermined significance) in thyroid cytology

“…11 Based on the results of published studies, the second (2017) edition of TBSRTC recommended subclassifying AUS/FLUS in descriptive language, including cytologic atypia (AUS-C), architectural atypia (AUS-A), cytologic and architectural atypia (AUS-C&A), Hürthle cell atypia (AUS-H), and atypia, not otherwise specified (AUS-NOS). 15 While new data on the ROM of these subclassifications of AUS/FLUS are still accumulating, [26][27][28][29][30] evidence of a significant difference in ROM between two major sub-groups, that is nuclear atypia and architectural atypia, is clear and overwhelming. The relatively wide ranges of documented ROM are attributed to many factors, T A B L E 1 Demographic and follow-up information of the patients with cytologic diagnosis of AUS and FLUS.…”

Risk stratification of indeterminate thyroid nodules by novel multigene testing: a study of Asians with a high risk of malignancy

“…Regardless, ThyroSeq involves a seven‐gene panel test with high positive predictive value (PPV) (97%) and is regarded as a ‘rule‐in’ test [ 14 ], suggesting that cytologically indeterminate nodules with positive molecular results should be managed for surgery. Recently, the latest ThyroSeq v3 version of the 112‐gene panel has increased the negative predictive value (NPV) (90.9–97%) of the test, and combined ‘rule‐in’ and ‘rule‐out’ uses, in line with the Afirma GSC platform, are being used [ 14 , 15 , 16 ]. Of note, most studies of the two popular molecular platforms, Afirma and ThyroSeq v3, are based on populations from North America with a low risk of malignancy (ROM) of 10–40% in indeterminate categories [ 11 , 12 ].…”