Cost‐effectiveness of Gene Expression Classifier Testing of Indeterminate Thyroid Nodules Utilizing a Real Cohort Comparator

Objective To predict the cost‐effectiveness of implementing routine gene expression classifier testing for thyroid nodules with indeterminate fine‐needle aspiration cytology, by utilizing a real cohort of patients as a comparator. Study Design Cost‐effectiveness analysis of a retrospective cohort compared with a simulated cohort. Setting Tertiary academic medical center. Subjects and Methods We reviewed the records of all patients who underwent ultrasound‐guided fine‐needle aspiration from 2010 to 2014 at a tertiary academic medical center. All patients with Bethesda class III or IV cytopathology had the details of their management catalogued over the subsequent 2 years of care. These patients were assigned to the standard‐of‐care arm of the study. We compared the third‐party payer costs of care and the proportion of patients who underwent surgery with a simulated cohort who underwent gene expression classifier testing after an initial indeterminate fine‐needle aspiration (molecular test arm). Results The cost of managing 1 nodule over 2 years was $2399 higher (range, $397‐$4399) for the molecular test group than the standard of care group. The molecular test group had a 13.1% decrease (base parameters; range, 0.73%‐45.09%) in the number of patients undergoing surgery. Conclusion When applied to a real cohort of patients, routine gene expression classifier is predicted to reduce the number of patients undergoing surgery but will increase cost of care. Cost‐effectiveness is heavily dependent on prevalence of malignancy and gene expression classifier specificity.