Evaluation of EORTC QLQ-C30 and QLQ-STO22 in a phase II, placebo-controlled trial of rilotumumab (R) plus epirubicin, cisplatin, and capecitabine (ECX) in patients (pts) with locally advanced (LA) or metastatic (m) gastric (G) or esophagogastric junction (EGJ) cancer.
e15046 Background: To select a patient-reported outcomes (PRO) instrument for use in a pivotal G/EGJ cancer study, we evaluated the characteristics of two quality of life (QoL) questionnaires in a phase II trial of R (a fully human monoclonal antibody that neutralizes HGF, the MET receptor ligand) or placebo + ECX in the first-line treatment of 121 pts with LA or mG/EGJ cancer. The trial showed trends for improved survival in the R arms, especially in pts with METHigh(> 50% cells +ve). Methods: Discriminant validity of EORTC QLQ-C30 and QLQ-STO22 was assessed by mean baseline (BL) scores grouped by ECOG PS 0 or ≥1. Responsiveness was assessed by longitudinal analyses of mean PRO score changes from BL by change in ECOG PS (better, no change, or worse). Results: Questionnaire completion was ≥ 60% for cycles 1-5. QLQ-C30 scores at BL had good discriminant validity by ECOG PS: ≥1 corresponded to lower QoL. QLQ-C30 was responsive to change from BL: higher mean QLQ-C30 scores were associated with improved ECOG PS. In contrast, QLQ-STO22 scores had poor discriminant validity and poor responsiveness to change from BL. Results were consistent across treatment arms, the intent-to-treat population, and MET expression groups. Conclusions: EORTC QLQ-C30, a global measure of QoL, correlates well with ECOG PS and is a valid and responsive QoL measure in this pt population. QLQ-STO22 is more specific for stomach-related symptoms and did not correlate with ECOG PS. QLQ-C30 will be included in a phase III, randomized, placebo-controlled trial of R + ECX for the first-line treatment of MET+ve LA or mG/EGJ cancer (NCT01697072). Clinical trial information: NCT00719550. [Table: see text]