Neoadjuvant therapy for localized gastrointestinal stromal tumors (GISTs): A propensity score weighted analysis.

Abstract

e23516 Background: Retrospective and single-arm prospective studies have reported clinical benefit associated with receipt of neoadjuvant imatinib for GISTs. In the absence of randomized phase III data, the impact of neoadjuvant systemic therapy (NAT) on survival, in comparison to upfront resection, remains unknown. Methods: We identified N = 14,402 patients with complete clinical, demographic, treatment and pathologic data within the National Cancer Database (2004-2016) who underwent resection of localized GIST of the stomach, esophagus, small bowel, and colorectum, with or without ≥3 months of NAT. Inverse probability of treatment weighting (IPTW) was used to adjust for covariable imbalance among treatment groups, with the propensity score estimated by logistic regression. The effect of NAT on overall survival was estimated with a weighted time-dependent Cox proportional hazards model. A weighted logistic regression was used to estimate the effect of NAT on 90-day postoperative mortality and R0 resection. Results: 759 (5.3%) patients received NAT followed by resection, compared to 13,643 (94.7%) who underwent upfront resection. Median length of NAT was 6.3 months. 53% of NAT patients were male vs. 49% of UR patients, 68% vs. 66% had primary gastric GIST, and 73% vs 49% were high risk. Patients receiving NAT had larger tumors (p < 0.001) and higher mitotic index (p = 0.003). There was a significant survival benefit associated with receipt of NAT (table). 90-day postoperative mortality rate was 3/759 (0.4%) among NAT patients vs. 307/13,643 (2.3%) UR patients. Receipt of NAT was significantly associated with lower odds of 90-day postoperative mortality (table). Of the 13,562 patients with information on margin status, the R0 resection rate was 635/716 (88.7%) for the neoadjuvant group vs. 11,823/12,846 (92%), with no significant difference between treatment groups (table). Conclusions: After adjustment for imbalance in prognostic and demographic factors, this analysis demonstrates that receipt of NAT for localized GIST is associated with a modest overall survival benefit. Although NAT patients had higher risk features, NAT was associated with a lower risk of 90-day postoperative mortality, with no difference in likelihood of achieving an R0 resection. [Table: see text]