Impact of Mutation Profile on Outcomes of Neoadjuvant Therapy in GIST

Abstract

Background: Neoadjuvant imatinib therapy plays a crucial role in the management of gastrointestinal stromal tumors (GISTs), but its impact across various mutational profiles remains uncertain.

Objective: The aim of this study is to describe the clinicopathological features and to assess the response and surgical outcomes of neoadjuvant imatinib in GIST patients exhibiting diverse mutational profiles.

Methods: We conducted a retrospective study, extracting data from the Dutch GIST Registry, including patients treated with neoadjuvant imatinib. Response rate was the primary outcome, and secondary outcomes were the time on neoadjuvant treatment and resection margins (R0 vs. R1/R2), respectively.

Results: Between 2009 and 2021, 326 patients were treated with neoadjuvant imatinib, of which 264 (80.9%) underwent resection. A total of 197 (74.6%) of them had a KIT-exon 11 mutation, 19 (7.3%) had other KIT mutations, 10 (3.8%) had PDGFRA D842 mutations, 21 (6.8%) had other PDGFRA mutations, 2 (0.7%) had NTRK mutation, 1 (0.4%) had an SDH mutation, and 17 (6.4%) had WT GISTs. Patients with KIT-exon 11 mutations demonstrated a higher rate of partial response to imatinib (60.5% vs. 33.3%; p = 0.00). A positive resection margin (R1 or R2) was observed in 14 (21.2%) patients with a non-KIT exon 11 mutations and in 11 (5.5%) patients with a KIT-exon 11 mutation (p = 0.00). Moreover, non-KIT exon 11 mutation patients had a shorter median duration of neoadjuvant therapy (5.3 months, range 0.5-21.0) compared to patients with a KIT exon 11 mutation (8.8 months, range 0.2-31.3; p < 0.001).

Conclusions: Our study highlights the variability in treatment response associated with different GIST mutational profiles. Patients with a KIT-exon-11 mutation tended to respond more favorably to neoadjuvant imatinib in terms of partial response and surgical outcomes.

Keywords: mutational profile in treatment of GIST; neoadjuvant imatinib in GIST.

Figure 1

Flowchart.

Figure 2

A Bland–Altman plot is provided to illustrate the agreement between tumor size assessed by preoperative CT scans and tumor size measured on resection specimens. In the plot, the x-axis represents the average of the two measurements (CT scan and resection specimen size), while the y-axis shows the difference between the two measurements for each patient. This Bland–Altman plot highlights the extent of variability between the two methods, and it is evident that there is a considerable range of differences in tumor size assessment across patients. The discrepancies between the tumor size measurements were observed in 28.2% of patients.