The benefit of cytoreductive surgery during imatinib (IM) therapy in patients with metastatic gastrointestinal stromal tumors (mGISTs): Retrospective analysis from several cancer centers.

Abstract

11523 Background: the role of cytoreductive surgery for patients with mGISTs during IM therapy has not been established yet because the absence of randomized trials. We carried out a retrospective analysis of the outcome of patients with mGISTs in several Moscow cancer centers. Methods: we compared several cohorts: treated with IM only (IM group) and treated with IM and surgery, the latter was divided into four groups depending on the time of the surgery and the response to IM: before IM (BIM), after partial response or stable disease (RD), or unifocal progressive disease (UPD) or multifocal progressive disease (MPD). Patients received IM after surgery until progression. The primary end points were progression free survival and overall survival from the start of IM therapy. Results: 234pts with mGISTs from 2002 till 2024 received IM 400 in the first line. 116 pts received imatinib only, 118 patients underwent cytoreductive surgery: before IM (BIM, n=39) and during IM – on responsive disease (RD, n=23), unifocal progressive disease (UPD, n=22), and multifocal progressive disease (MPD, n=27). 7 patients were excluded from the analysis because they underwent surgery twice (before and during IM). Cytoreductive surgery increased the median PFS in comparison with IM only (Table). The differences were statistically significant between IM vs BIM – 24.0 vs 61.0 (p=0.028), IM vs RD – 24.0 vs 77.0 (p=0.003), and RD vs MPD – 77.0 vs 39.0 (p=0.005). Patients with UPD and MDP had a median PFS after R0/R1 and R2 was 17.0 vs 7.0 (р=0.564) and 14.0 vs 7.0 (р=0.056) months, respectively. Patient who underwent surgery on progression received IM after surgery for 10.0 and 11.0 months in UPD and MPD groups respectively what was two times longer than on sunitinib therapy (the median PFS 6.8 months, NCT00075218). On multivariate analysis for the entire cohort radiologic response was predictive for PFS: RD (hazard ratio (HR) 0.37, p = 0.003), UPD (HR 0.47, p=0.03) and BIM (HR 0.48, P = 0.009) were independent prognostic factors of better PFS; KIT exon 9 mutations (HR 2.68, p=0.003) and unknown location of primary tumor – factors of worse PFS. Overall survival was reached only in MPD group – 75.0 (95% CI: 37.14-112.87) months from the start of IM. Conclusions: cytoreductive surgery during IM therapy in responsive disease increase PFS; metastasectomy on UPD is likely to be beneficial for patients with mGISTs and it can be more effective option than the second line treatment. PFS, median, months Standard error 95% CI Lower bound Upper bound IM only (n=116) 24.000 4.459 15.260 32.740 Surgery before IM (n=39) 61.000 28.061 6.001 115.999 RD (n=23) 77.000 21.954 33.971 120.029 UPD (n=22) 70.000 25,077 20.849 119.151 MPD (n=27) 39.000 5.747 27.736 50.264