OBJECTIVES: In the phase 3 METEOR trial (NCT01865747) cabozantinib significantly improved OS, PFS and ORR over everolimus in previously-treated patients. A priori between-treatment analysis showed cabozantinib maintained symptoms and HRQoL similarly to everolimus. Here, we report post-hoc within-treatment analysis of symptoms and impact on HRQoL.

METHODS: Patients completed FACT Kidney Symptom Index (FKSI-19) and EQ-5D-5L at baseline, every 4 weeks (first 6 months), then every 8 weeks and end of treatment (EOT). Repeated-measures regression of EQ-5D VAS on FKSI-19 symptoms was used to assess impact on HRQoL. Time until definitive deterioration (TUDD) using Kaplan-Meier/Cox methods was assessed for the FKSI-19 total and Disease-Related Symptoms 9-item subscale (FKSI-DRS9). Sensitivity analyses included a smaller minimal important difference (MID) and including disease progression and death as events. For time until FKSI-19 symptom impact, a patient report of “quite a bit/very much” was the event.

RESULTS: TUDD analyses included 615 (93.5%) ITT patients providing baseline and ≥1 post-baseline FKSI-19 value before EOT. Symptom analyses included 648 ITT patients providing ≥1 FKSI-19 value before EOT. Higher symptom frequency was associated with lower HRQoL for all FKSI-19 symptoms except ‘blood in urine’. No difference in time until FKSI-19 total deterioration was found between cabozantinib and everolimus; however, time until FKSI-DRS9 deterioration was longer for cabozantinib (5.6 vs 3.8 months, HR=0.79, 95%CI:0.64–0.99). Sensitivity analyses had consistent results and interpretation. Time until symptom impact was longer for most FKSI-19 symptoms (12/19) for cabozantinib versus everolimus but was shorter for diarrhea (HR=3.91, 95%CI:2.05-7.43). After post-hoc multiple testing adjustment, significant improvements remained with longer time to lack of energy (HR=0.50), fatigue (HR=0.57), pain (HR=0.47), shortness of breath (HR=0.26), coughing (HR=0.32), and HRQoL items relating to ability to work (HR=0.67) and contentment with HRQoL (HR=0.54).

CONCLUSIONS: Cabozantinib exhibited longer disease-related symptom control with an associated impact on quality of life.