Intermittent versus continuous administration of pazopanib in progressive radioiodine refractory thyroid carcinoma: Final results of the randomised, multicenter, open-label phase II trial PAZOTHYR.

ABSTRACT

INTRODUCTION: Multikinase inhibitor (MKI) treatments have shown efficacy in progressive radioiodine refractory thyroid cancers (RAIR-TC), but most patients experienced substantial adverse effects. This randomised multicentric study investigated intermittent versus continuous pazopanib administration. PATIENTS AND METHODS: The PAZOTHYR study included RAIR-TC patients with progressive disease in the last 12 months, who may have received one prior MKI. RAIR-TC patients received pazopanib for 6 months, and patients with stable disease or tumour response were randomly assigned (11) to receive continuous (CP) or intermittent (IP) pazopanib until progression. The primary end-point was time to treatment failure (TTF) defined as the time from randomisation to permanent discontinuation of pazopanib, due to any cause. One hundred randomised patients were needed to demonstrate an increase from 50% (CP) to 70% (IP) (hazard ratio (HR) 0.515, 80% power) in the rate of patients still under treatment 6 months (6m-SuT) post-randomisation. Secondary end-points included the overall response rate (ORR), progression-free survival (PFS) under pazopanib and safety. RESULTS: RAIR-TC patients (168) enrolled from June 18, 2013 to January 16, 2018, received 6-month pazopanib treatment and showed 35.6% (95% CI 28.2-43.6) best response rate and 89.4% (83.5-93.7) disease control rate. One hundred patients were randomised (IP50; CP50). With a median follow-up of 31.3 months, median TTF was not statistically different between arms (IP14.7, 95% confidence interval (CI) 9.3-17.4; CP11.9, 95% CI 7.5-15.6) months (HR 0.79, 0.49-1.27). 6m-SuT rates were similar (IP80% 66.0-88.7%; CP78% 63.8-87.2%). Median PFS under pazopanib were not statistically different (IP5.7 4.8-7.8; CP 9.2 7.3-11.1) months (HR 1.36, 0.88-2.12). Pazopanib-related adverse events grade 3-4 occurred in 36 (IP 19, 38%; CP 17, 34%) randomised patients. Seven pazopanib-related deaths occurred. CONCLUSIONS: Intermittent administration of pazopanib did not demonstrate significant superiority in efficacy or tolerance compared with continuous treatment. An intermittent administration scheme cannot be recommended outside clinical trials. This study was registered with ClinicalTrial.gov, number NCT01813136.