Predicting disease-specific survival in patients undergoing active surveillance for papillary thyroid carcinoma.

Abstract

Background: American Thyroid Association guidelines support active surveillance (AS) for low-risk papillary thyroid cancer (PTC). We developed a calculator to aid patient selection.

Methods: From 2004 to 2020, 148,904 PTC patients were selected from the surveillance, epidemiology, and end results (SEER) database. Univariable and multivariable analysis evaluated patient and treatment characteristics. Patients were randomly allocated into training (80%) or validation sets (20%). Coefficients generated a mathematical model to predict 5- and 10-year disease-specific survival (DSS).

Results: The mean DSS was 15.5 years with a 5- and 10-year DSS of 99.3% and 98.6%, respectively. Age, sex, race, median household income (MHI), tumor size, and nodal status were significant on multivariable analysis (p ≤ 0.05) and included variables in our calculator. 2404 patients underwent non-operative management (NOM) and were more likely older, male, higher MHI, larger tumor size, and less nodal positivity. Area under the curve (AUC) for 5- and 10-year DSS were 0.83 and 0.81, respectively, for the training set and 0.81 and 0.79, respectively, for the validation set.

Example: 65-year-old White female with a 0.8 cm PTC, cN0 with a MHI ≥ $75,000, had a 10-year predicted DSS was 95.6% with NOM and 99.3% with surgery. Alternatively, changing the patient's race to Hispanic, the 10-year predicted DSS was 94.1% with NOM and 99.0% with surgery.

Conclusions: As awareness of AS for PTC expands, it is important to consider objective data to guide informed decision making. This validated calculator is a useful tool to predict DSS for patients considering AS for PTC.