Development and Validation of mRNA Expression-Based Classifiers to Predict Low-Risk Thyroid Tumors

Allan Golding¹David Bimston¹Emma Namiranian²Ellen Marqusee²Gabriel Correa¹Evana Valenzuela-Scheker¹Ruochen Jiang³Yangyang Hao³Mohammed Alshalalfa³Jing Huang³Joshua P. Klopper^3*Richard T Kloos³Sara Ahmadi²

• ¹Memorial Healthcare System, Hollywood, Florida, United States
• ²Division of Endocrinology, Diabetes and Hypertension, Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts, United States
• ³Veracyte, South San Francisco, United States

Background: Molecular variants and fusions in thyroid nodules can provide prognostic information at a population level. However, thyroid cancers with the same molecular alterations exhibit diverse clinical behavior. Leveraging exome-enriched gene expression analysis may overcome the limitations of models based on a small number of point mutations or fusions. Here we developed and validated mRNA-based classifiers with high negative predictive values to preoperatively rule out thyroid tumor invasion and lymph node metastases. Materials and Methods: In this retrospective cohort study, histopathology reports from Afirma Genomic Sequencing Classifier (GSC) algorithm training and from consecutive thyroid cancer patients with Bethesda III-VI thyroid nodules in clinical practice (total 697 and ~50% from each) were scored for invasion and metastases. mRNA expression-based classifiers were developed utilizing literature-derived signatures as well as differentially expressed genes between samples with or without clinically significant invasion/metastases as the basic building blocks. Machine learning algorithms were employed to develop the final candidate classifiers. The final classifiers were validated on a retrospective cohort of 259 Afirma tested patients who had thyroid surgery and had invasion and metastasis scores assigned based on histopathology while blinded to the classifier results. Results: 697 (88% female) patient Afirma samples and scored histology reports were used for classifier development. In development, patients had a median age of 51 years. Ten percent of samples were assigned a high-risk for invasion label and 11.3% had were assigned a high-risk for lymph node metastases (LNM) label. A low-risk invasion classifier result was assigned to 41.3% of the cohort with a negative predictive value (NPV) of 97.6% and a low-risk LNM classifier result was assigned to 49.8% of the cohort with a NPV of 98.6%. In the validation cohort (75% female with a median age of 53 years), 51% of samples were ruled out for high-risk labeled invasion with a 99% [95-100] NPV and 53% were ruled out for high-risk labeled LNM with 100% [97-100] NPV. Discussion: Gene expression-based classifiers that confidently, preoperatively, rule out thyroid tumor invasion and lymph node metastasis may help personalize the surgical approach for individuals, reducing overtreatment, surgical complications, and postoperative hypothyroidism.