Small Bowel Gastrointestinal Stromal Tumors: A 15-Year Cohort Study Focusing on Jejuno-Ileal Site-Specific Outcomes...

Background: Site-specific long-term outcomes, including neurofibromatosis type 1 (NF1), Ki-67 prognostic value, and very late recurrences of small bowel gastrointestinal stromal tumors (GISTs), remain inadequately defined. Methods: This retrospective cohort study investigated the clinical characteristics, diagnostic challenges, and long-term outcomes of patients with small bowel GISTs. This retrospective, single-center study (2008-2024) analyzed 27 consecutive patients (average age: 62.2 years) with jejunal/ileal GISTs. Clinicopathologic features, diagnostic yield of balloon-assisted enteroscopy (BAE), treatments, and outcomes were evaluated during a 10.2-year median follow-up period. Recurrence-free survival (RFS) and overall survival (OS) were estimated by Kaplan-Meier with log-rank testing. Ki-67 was assessed using MIB-1; a prespecified 5% cut-off was chosen based on prior evidence. Results: Tumor (mean size, 62.4 mm) sites included the jejunum (74.1%) and ileum (25.9%). NF1 was present in 3/27 (11.1%) patients, all with multiple jejunal tumors. Among the 14 patients who underwent BAE, biopsy was attempted in six and yielded a histological diagnosis in one (16.7%). Six patients had recurrence; two died from disease >10 years postoperatively. Five-year OS and RFS were 91.3% and 68.7%, respectively. Adverse RFS was associated with ileal location (p = 0.03), size ≥ 10 cm (p < 0.001), mitoses > 5/50 high-power fields (p = 0.002), and Ki-67 ≥ 5% (p < 0.001). One patient labeled low risk by conventional models had recurrence with Ki-67 = 10%. Another classified as low risk by conventional models experienced recurrence >10 years after surgery, with a Ki-67 index of 10%. Conclusions: Extended, risk-adapted surveillance may be reasonable for small-bowel GISTs, and it may be beneficial to incorporate Ki-67 (≥5%) into site-based risk stratification. These observations remain hypothesis-generating and require validation in larger, multicenter cohorts and prospective studies.