Low Frequency of Mutation Testing in the United States: An Analysis of 3866 GIST Patients

The objective of this study was to determine whether there were survival differences associated with KIT mutation testing, type of KIT mutations, and other clinical variables in patients with localized or metastatic gastrointestinal tumor (GIST).

Adult patients with GIST were extracted from the Surveillance Epidemiology and End Results (SEER) database from 2010 to 2015 with follow-up through 2016. Overall survival (OS) and cancer-specific survival (CSS) were the designated endpoints.

A total of 3866 patients met inclusion criteria. Metastatic disease was found in 656 patients (17%), whereas localized disease was present in 3210 patients (83%). KIT mutation testing was performed in 1033 patients (26.7%) with equal distribution in localized and metastatic disease (27% and 26.6%, respectively). Multivariate analysis was performed in localized and metastatic GIST. In localized GIST, black race showed worse OS (hazard ratio [HR]=1.57; 95% confidence interval [CI]: 1.26-1.96), whereas higher mitotic rate (>5/50 HPF) demonstrated poor OS (HR=1.59; 95% CI: 1.24-2.05) and CSS (HR=3.07; 95% CI: 2.07-4.54); tumor size (>10 cm) showed poor CSS (HR=5.73; 95% CI: 2.37-13.8). In metastatic GIST, black race showed poor OS (HR=1.42; 95% CI: 1.04-1.93) and CSS (HR=1.73; 95% CI: 95% CI: 1.18-2.54), while KIT testing was associated with better OS (HR=0.64; 95% CI: 0.47-0.87) and CSS (HR=0.66; 95% CI: 0.44-0.97); treatment with tyrosine kinase inhibitors showed better OS (HR=0.67; 95% CI: 0.51-0.88). Surgical resection was associated with better OS (HR=0.56; 0.47-0.67) and CSS (HR=0.55; 95% CI: 0.42-0.72) both in localized and metastatic GIST.

The minority of GIST patients have their tumor tested for any KIT mutation. Yet, KIT testing and therapy with tyrosine kinase inhibitors were associated with better survival in GIST patients with metastatic disease. Surgery, potentially curative for localized GIST, shows benefit in the metastatic setting.