Validation of the EORTC QLQ-C30 Summary Score in patients with hematological malignancies.

Abstract

e18551 Background: The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) is a widely used questionnaire, which assesses 15 patients’ self-reported health domains, including physical and emotional functioning and key cancer symptoms. As this poses analytical challenges such as multiple testing, a single QLQ-C30 summary score (SS) was recently validated in solid tumors. In this work, we investigated its validity in patients with hematological malignancies. Methods: According to the published scoring algorithm, the SS is the mean of 13 QLQ-C30 scale scores, a higher SS rating reflecting a better health status. We performed univariate linear regression analysis to investigate the ability of the SS to discriminate between patients groups differing on known characteristics, i.e. ECOG performance status (0-1 vs. 2-3), comorbidity (yes vs. no), red blood cells transfusion dependency (TD, yes vs. no, MDS patients only), age (≤ 60 vs. > 60 years) and sex. We used linear mixed models to assess responsiveness of the SS to change, comparing baseline versus post-induction scores. For all analyses, we compared the SS performances to those of each original QLQ-C30 scale by Cohen's effect size (ES). Results: Analyses were based on pooled data from 2,134 patients diagnosed with myelodysplastic syndromes (MDS, n = 1,031), chronic myeloid leukemia (n = 382), acute myeloid (AML, n = 477) and promyelocytic leukemia (n = 244). Overall, 56.4% patients were male and the mean age was 51.5 years. Comparing groups by ECOG and comorbidity, SS outperformed 12 of the QLQ-C30 scales for ECOG (ES = 0.66) and 11 scales for comorbidity (ES = 0.17). With patients grouped by age (ES = 0.17) and sex (ES = 0.25), SS outperformed 13 and 12 scales respectively. For TD the SS (ES = 0.21) showed better discrimination than eight QLQ-C30 scales. When comparing baseline versus post-induction scores (AML patients), SS was more sensitive to change (ES = 0.41) than 10 of QLQ-C30 scales. Conclusions: Our findings provide support for the validity of the EORTC QLQ-C30 summary score as an endpoint in patients with hematological malignancies, when previous knowledge does not allow for domain-specific hypotheses.