PSYCHOMETRIC VALIDATION OF THE QLQ-AA/PNH-54 BASED ON A POOLED DATASET OF A CLINICAL TRIAL AND TWO REAL-WORLD STUDIES OF CROVALIMAB IN PARTICIPANTS WITH PAROXYSMAL NOCTURNAL HEMOGLOBINURIA
Author(s)
Jens Panse, MD1, Carla Mamolo, BSc, MASc, PhD2, Tracy Holt, PHD3;
1University Hospital RWTH Aachen; Centre for Integrated Oncology (CIO), Aachen, Germany, 2Genentech Inc., South San Francisco, CA, USA, 3Genentech Inc., a Member of the Roche Group, South San Francisco, CA, USA
1University Hospital RWTH Aachen; Centre for Integrated Oncology (CIO), Aachen, Germany, 2Genentech Inc., South San Francisco, CA, USA, 3Genentech Inc., a Member of the Roche Group, South San Francisco, CA, USA
OBJECTIVES: The QLQ-AA/PNH-54 is a patient-reported outcome developed to assess quality of life in patients with paroxysmal nocturnal hemoglobinuria (PNH) and acquired aplastic anemia (AA), two inter-related, ultra-rare hematological disorders. It was developed following the four-phase EORTC process and consists of 12 proposed domains. Here, the initial phase IV psychometric validation findings are presented.
METHODS: Data was pooled from 3 studies of adults with PNH who completed the QLQ-AA/PNH-54: Phase III COMMODORE 2 trial (N=23) of crovalimab (a novel anti-C5 monoclonal antibody approved for treatment of PNH), COMMODORE Burden of Illness real-world (RW) study (N=146), and PICNIC Health RW study (N=53). The QLQ-AA/PNH-54, EORTC QLQ-C30, and FACIT-Fatigue were administered at baseline in all studies. Item descriptive characteristics were assessed. Internal consistency and construct validity were assessed with Cronbach’s alpha and Spearman correlations, respectively. Standard deviations (SD) were calculated to provide a distribution-based estimate of meaningful change thresholds for each domain.
RESULTS: Of the 222 participants with baseline assessments, 114 (51.4%) were female; mean age was 41.9 (range 18-88) years. Four items demonstrated possible floor effects. Excellent internal consistency (Cronbach’s alpha >0.80) was observed for 8 domains; 3 domains (each <3 items) exhibited low internal consistency (<0.20). Moderate-to-high (0.5-0.8) inter-item correlations were observed for 8 domains, although Social Support, Body Image and Other Symptoms domains had little-to-low inter-item correlations (<0.20). Construct validity demonstrated moderate-to-high Spearman correlations (0.62-0.78) between most domains of the QLQ-AA/PNH with their QLQ-C30 counterparts; however, some correlations (e.g. Social Support with Social Functioning) were weaker than expected. Using 0.5SD, meaningful change thresholds for the domains ranged from 11-15.
CONCLUSIONS: QLQ-AA/PNH-54 demonstrated mostly strong psychometric properties based on a pooled dataset of 3 crovalimab studies of participants with PNH.
METHODS: Data was pooled from 3 studies of adults with PNH who completed the QLQ-AA/PNH-54: Phase III COMMODORE 2 trial (N=23) of crovalimab (a novel anti-C5 monoclonal antibody approved for treatment of PNH), COMMODORE Burden of Illness real-world (RW) study (N=146), and PICNIC Health RW study (N=53). The QLQ-AA/PNH-54, EORTC QLQ-C30, and FACIT-Fatigue were administered at baseline in all studies. Item descriptive characteristics were assessed. Internal consistency and construct validity were assessed with Cronbach’s alpha and Spearman correlations, respectively. Standard deviations (SD) were calculated to provide a distribution-based estimate of meaningful change thresholds for each domain.
RESULTS: Of the 222 participants with baseline assessments, 114 (51.4%) were female; mean age was 41.9 (range 18-88) years. Four items demonstrated possible floor effects. Excellent internal consistency (Cronbach’s alpha >0.80) was observed for 8 domains; 3 domains (each <3 items) exhibited low internal consistency (<0.20). Moderate-to-high (0.5-0.8) inter-item correlations were observed for 8 domains, although Social Support, Body Image and Other Symptoms domains had little-to-low inter-item correlations (<0.20). Construct validity demonstrated moderate-to-high Spearman correlations (0.62-0.78) between most domains of the QLQ-AA/PNH with their QLQ-C30 counterparts; however, some correlations (e.g. Social Support with Social Functioning) were weaker than expected. Using 0.5SD, meaningful change thresholds for the domains ranged from 11-15.
CONCLUSIONS: QLQ-AA/PNH-54 demonstrated mostly strong psychometric properties based on a pooled dataset of 3 crovalimab studies of participants with PNH.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
PCR190
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Rare & Orphan Diseases