2.1 Study population, data selection
This retrospective analysis relies on data of four multi-center LUX-Lung (LL) trials (LL1: NCT00656136, LL3: NCT00949650, LL5: NCT01085136, LL8: NCT01523587). Participants were non-small-cell lung cancer (NSCLC) patients with activating EGFR mutations (LL1 (36), LL3 (37, 38), LL5 (39)), or squamous cell carcinoma (LL8 (40)). The studies investigated the efficacy of Afatinib in comparison to placebo (LL1), chemotherapy (LL3, LL5), and Erlotinib (LL8). All LL trials collected HRQoL data using both the EORTC QLQ-C30 and the EQ-5D-3L.
Data for analyses was included if both HRQoL measures were completed at baseline and end of treatment (EOT). A flow chart of data selection is presented in Figure 1. We compared original study sample characteristics with selected data to check for potential selection bias, considering misalignment of more than 10% between the groups a signal for bias (41) (see Table 1).
2.2 Measures
EORTC QLQ-C30
The EORTC QLQ-C30 (26) consists of 30 questions that form 15 scales. Five of them are functioning scales (e. g. physical, cognitive); nine are symptom scales (e. g. pain, appetite loss). One scale covers global health status. Responses are provided on a four-point Likert scale (“not at all”, “a little”, “moderate”, “very much”) for all questions with the exception of the global health status items, which have seven response levels.
EORTC QLU-C10D
The QLU-C10D (24) is a preference-based scoring algorithm for the QLQ-C30 and enables health utilities to be derived from QLQ-C30 data. It uses 13 of the QLQ-C30 items, which form ten dimensions (physical functioning, role functioning, social functioning, emotional functioning, pain, fatigue, sleep, appetite, nausea, and bowel problems).
EQ-5D-3L
The EQ-5D-3L, developed by the EuroQol Group (16, 17), captures five dimensions – mobility, self-care, usual activities, pain/discomfort, anxiety/depression – with three response options each (“no problems”, “some problems”, “severe problems/unable to”) and is a validated instrument for a range of difference diseases, including cancer.
2.3 Statistical Analysis
Descriptive statistics are presented as absolute numbers (N/n = sample/group size, k = number of tests), central tendency (mean and/or median), standard deviation (SD), frequencies and percentages (%). Each data set of the four trials was analyzed separately. Additionally data of progressive patients of LL1 and LL8 were pooled for longitudinal analysis from baseline to progression date.
Country tariffs of both measures were calculated according to previously published algorithms (AUS (32, 42), CAN (33, 43), ITA (34, 44), NLD (35, 45), POL (34, 46), UK (36, 47)).
Clinically known groups, i.e. patient groups that were expected to differ with regard to HRQoL, were (48): ECOG PS (0 vs. > 0) for cross-sectional sensitivity analysis at baseline; three groupings (treatment arms, Adverse Events (AE; ≤ 3 vs. > 3), response (stable vs. progressive)) and pooled data of progressive patients for longitudinal responsiveness analysis.
Responsiveness to change was analyzed from baseline to EOT, within and between clinically known groups separately for each LL study. Also, change from baseline until disease progression was analyzed for the pooled data of progressive patients. We used independent and paired t-tests applying an α-level of 5%, and applying Bonferroni-Holm correction to adjust for multiple testing (49, 50). Further analysis only considered comparisons where at least one of the measures detected a significant difference.
Relative efficiency (RE) was used as a parameter for direct comparison of QLU-C10D and EQ-5D-3L utilities with regard to statistical efficiency (51). RE is calculated as the quotient of positive t-values of both measures, with the QLU-C10D in the numerator. Values larger than “1” indicate higher efficiency of the QLU-C10D, while values smaller than “1” indicate higher efficiency of the EQ-5D-3L.
Cohen’s d served as the effect size (ES) measure for sensitivity analysis, with scores ≥ 0.5 being considered moderate, and ≥ 0.8 being considered large (52). As effect size measure for change over time the Standardized Response Mean (SRM) was calculated by dividing mean change by the SD of the change (53). The Responsiveness Index (RI) was calculated to quantify change within clinically known sub-groups (54) as the quotient of mean change in the intervention/event group (intervention arm, > 3 AE, progressive patients) and the SD of the comparator group (control arm, ≤ 3 AE, stable patients). To compare both measures we calculated the differences between positive ES (QLU-C10D minus EQ-5D-3L). Differences > 0 favor the QLU-C10D and < 0 favor the EQ-5D-3L.