Afirma Xpression Atlas Alteration Status Is Associated with Malignancy Risk in Indeterminate Thyroid Nodules

Background: Molecular testing with the Afirma Genomic Sequencing Classifier (GSC) is commonly used to refine the diagnosis of thyroid nodules with indeterminate cytology. The Afirma Xpression Atlas (XA) is a whole-transcriptome RNA-sequencing platform that reports expressed alterations in GSC-suspicious nodules. The clinical implications of XA results have not been clearly defined. This study evaluated the association between XA alteration detection and histopathology outcomes. Methods: A retrospective review was performed of consecutive patients who underwent thyroid fine needle aspiration at a single institution from 3/2020 to 9/2025. All nodules with Bethesda III/IV cytology and GSC-suspicious results with subsequent XA testing that underwent surgical resection were included. Clinicopathologic variables were compared between XA alteration-positive and alteration-negative groups. Nodules were categorized as benign, noninvasive follicular thyroid neoplasm with papillary-like nuclear features, or malignant based on surgical pathology. Subgroup analyses were performed to compare clinicopathologic variables between BRAF -alteration, RAS -alteration, other-alteration, and alteration-negative groups. All analyses involving the other-alteration subgroup were exploratory. Multivariable linear regression was performed to identify independent predictors of 2025 American Thyroid Association (ATA) risk of recurrence category. Results: The study cohort included 192 patients (77.6% female) with a median age of 51 years. Among all 200 nodules, 94 nodules (47.0%) were alteration-positive. Nodule size did not differ between alteration-positive and alteration-negative groups. The malignancy rate was highest in BRAF -alteration nodules (100%; exact 95% Confidence Interval (CI): [71.5%, 100.0%]), followed by other-alteration (67.7%; 95% CI: [48.6%, 83.3%]), RAS -alteration (55.8%; 95% CI: [41.3%, 69.5%]), and alteration-negative nodules (46.2%; 95% CI: [36.5%, 56.2%]) ( p < 0.001). Despite a higher malignancy rate, alteration-positive nodules did not demonstrate more aggressive histopathologic features compared with alteration-negative nodules, including rates of extrathyroidal extension, lymphatic invasion, or angioinvasion. On multivariable linear regression, no alteration subgroups or clinical covariates were identified as independent predictors of 2025 ATA risk of recurrence category. Conclusions: Alteration detection by Afirma XA offers incremental malignancy risk stratification beyond GSC-suspicious classification alone in indeterminate thyroid nodules. The substantial malignancy rate in GSC-suspicious alteration-negative nodules, with histopathologic features similar to those of alteration-positive nodules, supports continued consideration of surgical resection in such cases.