The Joint Annual Scientific Meetings of the Endocrine Society of Australia and the Society for Reproductive Biology 2018

Taft Lecture (#83)

Susan Mandel 1
  1. University of Pennsylvania Health System, Pennsylvania, United States

Thyroid nodules are a common medical diagnosis, and are detected increasingly as incidental findings on imaging studies performed for other indications, rather than by physical examination.  Associated with increased diagnostic imaging and fine-needle aspiration biopsy rates (FNA), the incidence of thyroid cancer has risen over recent years, mostly attributable to the diagnosis of small (<2cm) papillary thyroid cancers.  Hence, this supports the contention that thyroid cancer may now be overdiagnosed, a patient dies with and not from the cancer, which drives the motivation for developing a rational approach to the management of thyroid nodules. Thyroid ultrasound (US) is fundamental for nodule evaluation.  Sonographic pattern systems have been developed by several organizations and are used to assign nodules into several categories, each associated with a defined cancer risk. Based upon the assigned risk, nodule size cutoffs for FNA are recommended.  Importantly, FNA is not required for a common pattern called spongiform which has a cancer risk of <2-3%.   Once FNA is performed, cytology results are reported using the Bethesda system and ~20% of specimens are diagnosed as indeterminate (Bethesda 3 or 4) with an associated 15-30% cancer risk.  The likelihood of cancer in a nodule with an indeterminate cytology diagnosis can be refined by both its sonographic appearance as well as by subsequent molecular testing of the specimen.  Lacking a known single driver mutation, the different malignant histologies are associated with various mutations, translocations and gene expression changes.  Hence, commercial tests reflect varied methodologies and testing targets.  In addition, the validation of a molecular test as a predictor of malignancy or benignity relies on the nodule’s baseline probability of cancer determined by the sonographic pattern and cytology diagnosis.  Therefore, real world application of molecular testing may demonstrate different predictive values than validation studies. The clinician is confronted with multiple diagnostic options that are rapidly evolving.