Multivariate regression analyses of data from a randomised, double-blind, placebo-controlled study confirm quality of life benefit of epoetin alfa in patients receiving non-platinum chemotherapy

Abstract

Cancer-related anaemia is associated with a wide spectrum of symptoms that can negatively affect quality of life. Because epoetin alfa has demonstrated efficacy in correcting cancer-related anaemia, the impact of this treatment on quality of life was evaluated in a multinational, randomised, double-blind, placebo-controlled trial in 375 anaemic cancer patients receiving non-platinum-based chemotherapy. The cancer-specific measures of quality of life included the general scale (FACT-G Total) and fatigue subscale (FACT-An Fatigue subscale) of the Functional Assessment of Cancer Therapy-Anaemia and the Cancer Linear Analogue Scales measuring energy, ability to do daily activities, and overall quality of life. These measures were also used to examine the relationship between haemoglobin levels and quality of life. Both univariate and multiple linear regression analyses of quality of life data were performed. Results of the univariate analysis have been reported previously. The a priori-planned multiple linear regression analysis, which accounted for the effects of disease progression and several other possibly confounding variables on quality of life, showed a significant advantage for epoetin alfa over placebo for the five scales (all, P<0.05), and confirmed the results of the univariate analysis. For cancer-specific measures, significant correlations were demonstrated between baseline haemoglobin and quality of life (r, range: 0.14-0.26, all P<0.05) and between change in haemoglobin and change in quality of life (r, range: 0.26-0.34, all P<0.01). These findings provide evidence that increasing haemoglobin levels by epoetin alfa administration can significantly improve cancer patients' quality of life.

Figure 1

Study schema.

Figure 2

Timing of QoL assessments. Each symbol represents a QoL assessment for a patient at the specified time in the study. The three separate symbols represent the last assessment for a patient who withdrew early (star), the last assessment for a completer (closed circle), or a continuing assessment (open circle).

Figure 3

Flow diagram of the patients' progress through the phases of the trial.

Figure 4

QoL mean change scores by treatment group: results of univariate (Littlewood et al, 2001) and multiple linear regression analyses* (FACT-G Total scale and FACT-Fatigue Subscale). *P values adjusted for multiple comparisons (sequentially rejective Bonferroni procedure).

Figure 5

QoL mean change scores by treatment group: results of univariate (Littlewood et al, 2001) and multiple linear regression analyses* (CLAS scales). *P values adjusted for multiple comparisons (sequentially rejective Bonferroni procedure).

Figure 6

QoL mean change scores by treatment group: results of univariate (Littlewood et al, 2001) and multiple linear regression analyses* (SF-36 Summary scores). *P values adjusted for multiple comparisons (sequentially rejective Bonferroni procedure).

Figure 7

Visual graphic demonstrating the relationship between Hb and QoL over time by showing biweekly Hb levels and CLAS Energy levels from univariate analysis (Littlewood et al, 2001).