The placebo response in trials of drug treatments for cancer-related fatigue: A systematic review, meta-analysis and meta-regression


Session type:


Maria del Rocio Roji Buqueras1,Patrick Stone2,Federico Ricciardi2,Bridget Candy2
1University College London, London, UK,2The Marie Curie Palliative Care Research Department, University College London, London, UK



Cancer related fatigue (CRF) is one of the most distressing symptoms experienced by cancer patients. There is no gold standard treatment, although multiple drugs have been tested with little evidence of efficacy. Many randomized clinical trials (RCTs) reports in this area commented on the existence of a placebo response (PR). The objective of this systematic review was to establish the magnitude of the PR in RCTs of drugs to relieve CRF and to identify contributing factors.


We conducted a systematic review of RCTs comparing drug against placebo, in which the main objective was to treat CRF, in adult cancer patients, at any disease stage, where fatigue was measured using a validated tool. We conducted a meta-analysis using the standardized mean change (SMC) between baseline and final measurement in the placebo group. We undertook a meta-regression to explore factors that may have been associated with the PR (e.g. patient population or trial design). We assessed the risk of bias for each trial using the criteria outlined in the Revised Cochrane tool.


We identified 3916 publications of which 23 were included in quantitative analysis. Studies had limitations, which increased potential risk of bias. The pooled SMC estimate was -0.23 (95% CI -0.42 to -0.04). This is a moderate effect size that was statistically significant (p = 0.02). None of the multiple variables analysed in the meta-regression were statistically significantly related to PR.


There is some evidence, based on trials with small samples, that the PR in trials testing drugs for cancer related fatigue is moderate but statistically significant. No factors related to PR were identified. Since the PR is relatively large, we recommend that, in uncontrolled studies, potential treatments for CRF should be able to demonstrate an effect size of at least 0.23 before being considered for further evaluation in RCTs.