Utility and costs of surveillance imaging for low- and very low-risk gastrointestinal stromal tumors.

Abstract

There is limited evidence on the optimal surveillance approach in patients with low- and very low-risk gastrointestinal stromal tumors, resulting in inconsistent and arbitrary approaches to surveillance in this population. In this study, we reviewed our institutional approach to surveillance in patients with low- and very low-risk gastrointestinal stromal tumors and the costs associated with detecting recurrence. We retrospectively reviewed consecutive adult patients treated for low- and very low-risk gastrointestinal stromal tumors at our institution from 2010 to 2019. Data collected included patient and tumor characteristics, surgical management, and postoperative follow-up. Surveillance-related expenses were calculated using estimates of average costs obtained from our institution. A cost analysis was performed to evaluate estimated yearly costs based on the surveillance strategy used. There were 60 patients included. The mean age at diagnosis was 63.9 (±12.5) years. The primary tumor was typically in the stomach (73%; n = 44). Computed tomography scan of the abdomen and pelvis with intravenous contrast was the most common surveillance modality (total = 226 scans). No recurrences were identified. Median follow-up duration was 49.0 (interquartile range = 19.5-61.5) months. The mean number of surveillance images per patient was 4 (±2.6). Surveillance imaging was obtained more frequently than just annually in 83% (n = 50) of patients, with an estimated yearly cost of $2,840.77 (interquartile range = $2,273.62-$3,895.92) and no detection of recurrence. In this study population, patients with low- and very low-risk gastrointestinal stromal tumors underwent frequent imaging studies for surveillance with little yield and at substantial cost. Further multi-institutional studies on practice patterns and outcomes of surveillance are warranted to better inform standardized surveillance recommendations.