Patel, Kepal N1; Angell, Trevor E
2; Barbiarz, Joshua
3; Barth, Neil M
4,5; Blevins, Thomas C
6; Duh, Quan-Yang
7; Ghossein, Ronald A
8; Harrell, R Mack
9; Huang, Jing
3; Imtiaz, Urooj
5; Kennedy, Giulia C
3; Kim, Su Yeon
3; Kloos, Richard T
4; LiVolsi, Virginia A
10; Randolph, Gregory W
11; Sadow, Peter M
12; Shanik, Michael H
13; Sosa, Julie A
14; Traweek, S Thomas
15; Walsh, P Sean
3; Whitney, Duncan
3; Yeh, Michael W
16; Ladenson, Paul W
17
1 Division of Endocrine Surgery, Department of Otolaryngology-Head and Neck Surgery, NYU Langone Medical Center, NY, USA
2 Division of Endocrinology, Diabetes, and Hypertension; Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, USA
3 Department of Research & Development, Veracyte, Inc., South San Francisco, USA
4 Department of Medical Affairs, Veracyte, Inc., South San Francisco, USA
5 Department of Clinical Affairs, Veracyte, Inc., South San Francisco, USA
6 Texas Diabetes and Endocrinology, Austin, TX, USA
7 Section of Endocrine Surgery, Department of Surgery, University of California San Francisco, San Francisco, CA, USA
8 Division of Head and Neck Pathology, Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
9 The Memorial Center for Integrative Endocrine Surgery; Hollywood, Weston and Boca Raton, FL, USA
10 Anatomic Pathology Division, Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA, USA
11 Division of Thyroid and Parathyroid Endocrine Surgery, Department of Otolaryngology, Massachusetts Eye and Ear Infirmary and Harvard Medical School, Boston MA, USA
12 Head and Neck Pathology Subspecialty, Department of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston MA, USA
13 Endocrine Associates of Long Island, Smithtown, NY, USA;
14 Section of Endocrine Surgery, Department of Surgery, Duke University Medical Center, Durham, NC, USA
15 Thyroid Cytopathology Partners, Austin, TX, USA
16 Endocrine Surgery Program, Department of Surgery, UCLA David Geffen School of Medicine, Los Angeles, CA, USA
17 Division of Endocrinology, Diabetes and Metabolism; Department of Medicine; Johns Hopkins University, Baltimore, MD, USA
Background/ Purpose: We used next-generation RNA sequencing and advanced machine-learning algorithms to improve classification of cytologically indeterminate thyroid nodules.
Methods: Fine needle aspiration (FNA) specimens were collected preoperatively from 634 Bethesda III and IV nodules with benign and malignant labels and subjected to RNA Sequencing. Several algorithms were trained using a combination of differentially expressed genes, variants, fusions, mitochondrial genes, and loss of heterozygosity as input features into classification models. The best model, named the Genomic Sequencing Classifier (GSC), uses an Ensemble classifier of 12 sub-models to make a benign versus malignant prediction, as well as additional classifiers for follicular cell adequacy, Hürthle cell content and neoplasia, BRAFV600E mutation status, medullary thyroid carcinoma, and parathyroid tissue. Also included is a calling algorithm for RET/PTC fusions. We tested performance of the GSC on a previously described multi-center prospective cohort
1 of 210 Bethesda III and IV nodules, of which 191 had remaining RNA available for analysis, and an independent cohort of 324 consecutive indeterminate patient samples from the CLIA lab for which GEC results were known.
Results: The locked GSC demonstrated sensitivity=91%/specificity=68%, and at 24% cancer prevalence PPV=47%/NPV=96%. Among 324 consecutive CLIA specimens, >90% of GEC benign specimens were also GSC benign.
Discussion & Conclusion: The GSC demonstrated high sensitivity and >30% relative increase in specificity compared to the original Afirma GEC
1. GSC should significantly increase the number of patients identified with benign nodules and spare more unnecessary surgery, resulting in a greater proportion of patients undergoing surgery for malignancies.
References:
- Alexander EK, et al. Preoperative Diagnosis of Benign Thyroid Nodules with Indeterminate Cytology. N Engl J Med. 2012;367(8):705-15.