Adjuvant Imatinib or Observation in Patients With Gastrointestinal Stromal Tumors With KIT Exon 9 Mutations.

Importance: Gastrointestinal stromal tumors (GISTs) harboring KIT exon 9 mutations represent a biologically distinct subgroup with reduced sensitivity to standard-dose imatinib in the advanced setting. The benefit of adjuvant imatinib in this population remains uncertain. Objective: To evaluate the association between adjuvant imatinib and recurrence-free survival (RFS) and overall survival (OS) in patients with resected GISTs with KIT exon 9 mutations. Design, Setting, and Participants: This international, multicenter cohort study included patients with localized, molecularly confirmed GISTs with KIT exon 9 mutations who underwent curative-intent surgery between January 1990 and July 2022 at 35 referral centers in Europe, the US, and Japan, or were registered in the Life Raft Group database. The analysis took place between January 2025 and November 2025. Exposures: Adjuvant imatinib initiated after curative surgery, modeled as a time-dependent covariate to account for immortal time bias. Main Outcomes and Measures: The primary end points were RFS (time from surgery to recurrence or death) and OS (time from surgery to death) in the full cohort and in the high-risk subgroup defined by modified National Institutes of Health (mNIH) criteria. Multivariable Cox regression models included established prognostic covariates and cluster-robust standard errors. Overlap weighting (OW) based on propensity scores was used as a causal inference model. Secondary analyses evaluated 400 mg/d vs 800 mg/d dosing in the mNIH high-risk subgroup. Results: A total of 367 patients were included, 187 (51.0%) male and 180 (49%) female, with a mean (SD) age of 56 (13) years. Among these, 91 (24.8%) were observed and 276 (75.2%) received adjuvant imatinib (median [IQR] duration, 27.3 [13.5-36.0] months; 116 [42.0%] treated ≥3 years). Consistent with a cytostatic activity, adjuvant imatinib in the full cohort was associated with a reduced early hazard of recurrence or death (HR, 0.19; 95% CI, 0.10-0.36), with attenuation over time (time-interaction hazard ratio [HR], 1.85 per log-year). Treatment was also associated with improved OS (HR, 0.37; 95% CI, 0.17-0.83). Similar results were obtained when limiting the analysis to patients with mNIH high-risk disease. OW models and sensitivity analyses confirmed similar associations with RFS and OS. Among 257 patients with mNIH high-risk disease receiving imatinib, no significant difference was observed between 400 mg/d and 800 mg/d dosing. Conclusion and relevance: In this large international cohort study of resected GISTs with KIT exon 9 mutations, adjuvant imatinib was independently associated with delayed recurrence and improved survival. These findings support the use of adjuvant imatinib in mNIH high-risk GISTs with KIT exon 9 mutations and underscore the need for prospective studies to define optimal dosing and treatment duration.