The clinical utility of "suspicious" molecular profile results and the Afirma Xpression Atlas in detecting thyroid malignancy: A long-term single-center validation study of the Afirma Genomic Sequencing Classifier.

Authors

Justin Bauzon, Eduardo Canalizo, Amy Han, Angela Thelen, Ludovico Sehnem, Katherine Heiden, Judy Jin, Joyce Shin, Eren Berber, Vikram Krishnamurthy, Allan Siperstein, Gustavo Romero-Velez

Published in

Surgery. Pages 110654. Sep 16, 2026. Epub Sep 16, 2026.

Abstract

A "benign" result from the Afirma Genomic Sequencing Classifier molecular test effectively rules out malignancy in cytologically indeterminate thyroid nodules. When a nodule is interpreted as "suspicious," the ability of the Genomic Sequencing Classifier to predict cancer has been less precise. Separately, the Xpression Atlas panel identifies gene mutations that may augment Genomic Sequencing Classifier-suspicious results. We assessed the long-term performance of the Genomic Sequencing Classifier with Xpression Atlas to identify malignancy in indeterminate thyroid nodules.
This is a single-center retrospective analysis of indeterminate (former Bethesda III/IV), Genomic Sequencing Classifier-suspicious thyroid nodules that underwent surgical resection from July 2017 to September 2024. Nodules were classified as thyroid cancer or nonthyroid cancer based on final pathology. Outcomes included the positive predictive value of the Genomic Sequencing Classifier with and without an Xpression Atlas-identified mutation.
Of 593 resected Genomic Sequencing Classifier-suspicious nodules, 512 (86%) were malignant. Mutations were more frequently detected via Xpression Atlas in malignant nodules (34.4% vs 11.1%; P < .001) compared with the nonthyroid cancer group. The overall rate of malignancy was 86.3%, with an increasing positive predictive value from 2017 (83.3%) to 2024 (89.0%). Absolute malignancy risk was significantly lower with age >55 years (83.2% vs 89.7%; P = .021) and higher with an Xpression Atlas-identified mutation (95.1% vs 82.4%; P < .001). On multivariable analysis, the presence of any mutation independently predicted thyroid cancer (odds ratio, 4.03; 95% confidence interval, 2.06-8.9; P < .001).
In this large cohort study, the Afirma Genomic Sequencing Classifier demonstrated excellent rule-in performance, contrary to previous reports. Xpression Atlas-identified mutations further enhanced cancer detection, regardless of the specific mutation detected, supporting the clinical utility of Xpression Atlas with the Genomic Sequencing Classifier to correctly identify malignancy in indeterminate nodules.

PMID:
42859092
Bibliographic data and abstract were imported from PubMed on 11 Oct 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.