313 Differential Molecular Testing Signatures Between Atypia of Undetermined Significance (AUS)-Nuclear versus AUS-Other Categories: Correlation of cytopathologic features with TIRADS and Surgical Outcomes

Document Type

Conference Proceeding

Publication Date

3-23-2026

Publication Title

Lab Invest

Keywords

adult, aged, anaplastic carcinoma, cohort analysis, conference abstract, controlled study, cytopathology, diagnosis, female, human, human tissue, major clinical study, retrospective study, thyroid imaging reporting and data system

Abstract

Disclosures: Sarah Kisha: None; Ali Rizk: None; Kerem Ozcan: None; Lisi Yuan: None Background: In 2023, the Bethesda System for Reporting Thyroid Cytopathology recommended subclassifying Atypia of Undetermined Significance (AUS) into AUS-nuclear (AUS-N) and AUS-other (AUS-O). Molecular testing is recommended to help improve risk stratification and enhance clinical guidance. In this study, we are comparing the cytologic findings, Afirma molecular testing results, ultrasound risk stratification (TIRADS), and surgical pathology correlation between AUS-O and AUS-N with special emphasis on cytologic features. By analyzing these parameters, we are aiming to assess the performance of the new AUS subclassification platform and whether certain cytologic features can help predict the outcome of molecular testing. Design: We retrospectively analyzed 322 AUS cases in the period from September 2024 to August 2025, including AUS-O (n=245) and AUS-n (n=77). We then reviewed cytologic features, molecular testing results (Afirma), TIRADS, and the final surgical pathology. Results: Out of 322 AUS, 224 had Afirma testing results, including 173 AUS-O and 51 AUS-N. In AUS-O group (n=173), 130 were benign on Afirma testing; 43 (24.9%) were suspicious on Afirma testing, with 70% being RAS mutations. In AUS-N group (n=51), 32 were benign on Afirma testing; 19 (35.8%) were suspicious on Afirma testing. In AUS-O group, the majority of cases had TIRADS score of 4, regardless of Afirma results (48% vs 51%). In AUS-N group, the majority of cases with benign Afirma testing results had TIRADS score of 3 (41.9%); however, the majority of cases with suspicious Afirma testing results had TIRADS score of 4 (61.1%). 17 out of 43 (39.5%) AUS-O with suspicious Afirma testing results were resected, including 7 carcinomas (41.2%). In contrast, 11 out of 19 AUS-N (57.9%) with suspicious Afirma testing results were resected, including 6 carcinomas and 1 anaplastic carcinoma (63.7%). In AUS-O, review of cytologic features revealed that microfollicles, syncytial pattern, nuclear enlargement were more prevalent in Afirma suspicious group. In AUS-N, architectural atypia, nuclear enlargement, groove, and pseudoinclusions were more frequently seen in Afirma suspicious group. [Formula presented] [Formula presented] Conclusions: Our results support the new AUS subclassification, with AUS-N had more suspicious Afirma testing results. In AUS-N group, cases tested suspicious on Afirma had higher TIRADS scores and more AUS-N were resected with worse outcomes.

Volume

106

Issue

3

Share

COinS