Estimating Meaningful Change Thresholds (MCTs) for EORTC Scales in a Phase 3 Trial in Participants With Inoperable or Metastatic Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Breast Cancer

OBJECTIVES: Interpretation of clinical trial data is limited without defined MCTs for within-participant change and between-group difference of study endpoints. This study aimed to estimate MCTs for selected European Organisation for Research and Treatment of Cancer (EORTC) scales from the TROPION-Breast01 trial of advanced breast cancer (NCT05104866).

METHODS: Pre-specified analyses were performed to define MCTs for selected EORTC scales, including Global Health Status/Quality of Life (GHS/QoL), functioning (physical, role, emotional, cognitive, social), pain, fatigue, arm, and breast symptoms using pooled blinded data from baseline, weeks 6 and 12. Patient Global Impression of Severity and Patient Global Impression of Change were used as anchors. Anchor appropriateness was assessed via Spearman correlations, with values ≥0.371 considered adequate. Distribution and anchor-based methods were investigated. Test-retest and internal consistency reliability coefficients, required as input for the distribution-based methods, were examined. The within-participant MCT estimates were additionally evaluated against the possible amount of change observable on each scale.

RESULTS: In total, 513 and 384 participants provided evaluable scores at baseline and across all timepoints, respectively. Anchor correlations were low (<0.371) for majority of scales, thus anchor-based estimates were given limited consideration. Thresholds were estimated via distribution-based approaches, supported by adequate reliability for most scales. Within-participant threshold for deterioration was 11.1 for fatigue and arm symptoms, 13.3 for physical functioning, and 16.6 for GHS/QoL, functioning (role, emotional, cognitive, social), pain, and breast symptoms. GHS/QoL, physical functioning, and pain were employed as trial secondary endpoints via time to deterioration analyses. The between-group difference MCT range was 6.6–15.6.

CONCLUSIONS: MCTs were derived for the selected EORTC scales and helped Patient-Reported Outcome (PRO) endpoint interpretation in this trial. MCTs may also be used in routine care to better understand the significance of longitudinal change in patient outcomes.