Preoperative Identification of Medullary Thyroid Carcinoma (MTC): Clinical Validation of the Afirma MTC RNA-Sequencing Classifier

Abstract

Background: Cytopathological evaluation of thyroid fine-needle aspiration biopsy (FNAB) specimens can fail to raise preoperative suspicion of medullary thyroid carcinoma (MTC). The Afirma RNA-sequencing MTC classifier identifies MTC among FNA samples that are cytologically indeterminate, suspicious, or malignant (Bethesda categories III-VI). In this study we report the development and clinical performance of this MTC classifier. Methods: Algorithm training was performed with a set of 483 FNAB specimens (21 MTC and 462 non-MTC). A support vector machine classifier was developed using 108 differentially expressed genes, which includes the 5 genes in the prior Afirma microarray-based MTC cassette. Results: The final MTC classifier was blindly tested on 211 preoperative FNAB specimens with subsequent surgical pathology, including 21 MTC and 190 non-MTC specimens from benign and malignant thyroid nodules independent from those used in training. The classifier had 100% sensitivity (21/21 MTC FNAB specimens correctly called positive; 95% confidence interval [CI] = 83.9-100%) and 100% specificity (190/190 non-MTC FNAs correctly called negative; CI = 98.1-100%). All positive samples had pathological confirmation of MTC, while all negative samples were negative for MTC on surgical pathology. Conclusions: The RNA-sequencing MTC classifier accurately identified MTC from preoperative thyroid nodule FNAB specimens in an independent validation cohort. This identification may facilitate an MTC-specific preoperative evaluation and resulting treatment.

Keywords: indeterminate cytology; machine learning; medullary thyroid cancer; molecular diagnostics; molecular testing; thyroid nodule.

FIG. 1.

Flow diagram of the RNA-sequencing MTC classifier development and blinded independent validation. MTC, medullary thyroid carcinoma; SVM, support vector machine.

FIG. 2.

Participant flow diagram. The training cohort was derived from several IRB-approved FNAB sample collection protocols. The protocol for the 461-sample cohort was described in the supplement of Patel et al (see ENHANCE Arm 1, ENHANCE Arm 2, and CLIA-GEC B). These samples were supplemented with four additional FNAB samples that underwent microarray-based genomic testing in the Veracyte CLIA laboratory and resulted positive by the parathyroid classifier with subsequent surgical confirmation of parathyroid adenoma. The protocol to collect FNAB samples that underwent microarray-based genomic testing in the Veracyte CLIA laboratory and resulted positive by the Medullary Thyroid Cancer Classifier with subsequent surgical confirmation was previously described in Kloos et al. These samples were randomly assigned here to the training or validation cohorts. The 191 Bethesda categories III/IV samples in the validation cohort here were originally collected under a prospective multicenter blinded sample collection protocol for the Afirma GEC validation as described in Alexander et al. Collection and attribution of those samples are shown in that publication's Supplementary Figure 1. Those with sufficient residual RNA were then used to validate the Afirma GSC as described in Patel et al, its Supplementary Figure 2 and are the same samples for this validation. Similarly, the one additional FNAB sample added to this validation cohort was a Bethesda category V sample originally collected and described in the Afirma GEC validation cohort with MTC surgical pathology. It too met inclusion criteria for the Afirma GSC validation secondary test set. CLIA, Clinical Laboratory Improvement Amendments; FNAB, fine-needle aspiration biopsy; GEC, Gene Expression Classifier; GSC, Gene Sequencing Classifier.