Figure 1

CONSORT diagram.^aPatients who were neither randomised nor had discontinued treatment during the lead-in stage are shown in the open-label continuation group, including five patients with SD at week 12 based on final data analysis. Randomisation was halted early by recommendation of the study oversight committee due to the high rates of tumour regression and the observation of symptomatic progression in individual patients randomised to placebo in several of the disease cohorts. ^bOne patient originally randomised to cabozantinib remained active at the time of the data cutoff. ^cTwelve patients were randomly assigned to placebo and crossed over to open-label cabozantinib after unblinding. Four of those 12 patients remained active at the time of the data cutoff, and eight discontinued treatment (one patient request, one death, and six PD). CONSORT=Consolidated Standards of Reporting Trials; PD=progressive disease.