Background: Resistance to approved inhibitors of KIT proto-oncogene, receptor tyrosine kinase (KIT), and platelet-derived growth factor receptor α (PDGFRA) is a clinical challenge for patients with advanced gastrointestinal stromal tumours. We compared the efficacy and safety of ripretinib, a switch-control tyrosine kinase inhibitor active against a broad spectrum of KIT and PDGFRA mutations, with placebo in patients with previously treated, advanced gastrointestinal stromal tumours.

Methods: In this double-blind, randomised, placebo-controlled, phase 3 study, we enrolled adult patients in 29 specialised hospitals in 12 countries. We included patients aged 18 years or older who had advanced gastrointestinal stromal tumours with progression on at least imatinib, sunitinib, and regorafenib or documented intolerance to any of these treatments despite dose modifications, and who had an Eastern Cooperative Oncology Group (ECOG) performance status of 0-2. Eligible patients were randomly assigned (2:1) to receive either oral ripretinib 150 mg once daily (ripretenib group) or placebo once daily (placebo group). Randomisation was done via an interactive response system using randomly permuted block sizes of six and stratified according to number of previous therapies and ECOG performance status. Patients, investigators, research staff, and the sponsor study team were masked to a patient's treatment allocation until the blinded independent central review (BICR) showed progressive disease for the patient. The primary endpoint was progression-free survival, assessed by BICR. The primary analysis was done in the intention-to-treat population and safety was assessed in patients who received at least one dose of study drug. Patients randomly assigned to placebo were permitted to cross over to ripretinib 150 mg at the time of disease progression. The INVICTUS study is registered with ClinicalTrials.gov, number NCT03353753, and with WHO International Clinical Trials Registry Platform, number EUCTR2017-002446-76-ES; follow-up is ongoing.

Findings: Between Feb 27, 2018, and Nov 16, 2018, 129 of 154 assessed patients were randomly assigned to receive either ripretinib (n=85) or placebo (n=44). At data cutoff (May 31, 2019), at a median follow-up of 6·3 months (IQR 3·2-8·2) in the ripretinib group and 1·6 months (1·1-2·7) in the placebo group, 51 patients in the ripretinib group and 37 in the placebo group had had progression-free survival events. In the double-blind period, median progression-free survival was 6·3 months (95% CI 4·6-6·9) with ripretinib compared with 1·0 months (0·9-1·7) with placebo (hazard ratio 0·15, 95% CI 0·09-0·25; p<0·0001). The most common (>2%) grade 3 or 4 treatment-related treatment-emergent adverse events in the ripretinib group (n=85) included lipase increase (four [5%]), hypertension (three [4%]), fatigue (two [2%]), and hypophosphataemia (two (2%]); in the placebo group (n=43), the most common (>2%) grade 3 or 4 treatment-related treatment-emergent adverse events were anaemia (three [7%]), fatigue (one [2%]), diarrhoea (one [2%]), decreased appetite (one [2%]), dehydration (one [2%]), hyperkalaemia (one [2%]), acute kidney injury (one [2%]), and pulmonary oedema (one [2%]). Treatment-related serious adverse events were reported in eight (9%) of 85 patients who received ripretinib and three (7%) of 43 patients who received placebo. Treatment-related deaths occurred in one patient in the placebo group (septic shock and pulmonary oedema) and one patient in the ripretinib group (cause of death unknown; the patient died during sleep).

Interpretation: Ripretinib significantly improved median progression-free survival compared with placebo and had an acceptable safety profile in patients with advanced gastrointestinal stromal tumours who were resistant to approved treatments.

Funding: Deciphera Pharmaceuticals.