OBJECTIVES: Gastrointestinal stromal tumours (GISTs) are rare cancers with high recurrence rates after surgery. Imatinib (a tyrosine kinase inhibitor, TKI) delays progression but tumour resistance can require patients to move to second and then third line alternative TKIs (typically sunitinib and regorafenib). Few studies have compared newly emerging third line TKIs for efficacy.

METHODS: A systematic review (SR) applied a defined PICOS protocol to identify eligible RCTs, which were analysed for risk of bias. A meta-analysis compared median progression-free survival (mPFS) relative to placebo or best supportive care (BSC), using the generic inverse variance technique to generate fixed effects (FE) and random effects (RE) models.

RESULTS: The SR identified over 500 articles with five RCTs meeting the selection criteria for third line therapy, while also reporting mPFS hazard ratios. The pooled data in the RE model suggested all TKIs were favoured versus placebo/BSC for third line GIST treatment (HR of 0.43; 95% CI 0.24, 0.77; p= 0.005) but with a high I (86%) indicating major heterogeneity. Two studies with regorafenib had the lowest HR (HR 0.08, CI 0.02, 0.38 and HR 0.27, CI 0.19, 0.39), though imatinib re-challenge (HR 0.46, CI 0.27,0.77) and pazopanib (HR 0.59, CI 0.37, 0.95) were also favoured versus placebo/BSC. Nilotinib showed a low and non-significant reduction in risk (HR 0.90, CI 0.65, 1.25).

CONCLUSIONS: Results suggest patients with post-surgical advanced GISTS progressing after imatinib and sunitinib therapy can potentially benefit from third-line TKIs in terms of PFS. This applies to regorafenib (the recommended third line treatment) but also for imatinib re-challenge and pazopanib, but for nilotinib to only a very limited extent. Several other factors, not least patient treatment history and mutation profile, drug adverse events and overall survival outcomes need further examination to support choice of appropriate therapy.