Drug therapy for the management of cancer related fatigue

Background Cancer related fatigue (CRF) is common, under-recognised and difficult to treat. There have been trials looking at drug interventions to improve CRF but results have been conflicting depending on the population studied and outcome measures used. No previous reviews of this topic have been exhaustive or have synthesised all available data. Objectives To assess the efficacy of drugs for the management of CRF. Search strategy We searched the Cochrane Central Register of Controlled Trials (1st Quarter 2007), MEDLINE ( 1966 to March 2007) and a selection of cancer journals. We searched references of identified articles and contacted authors to obtain unreported data. Selection criteria Trials were included in the review if they 1) assessed drug therapy for the management of CRF compared to placebo, usual care or a non-pharmacological intervention in 2) randomised controlled trials (RCT) of 3) adult patients with a clinical diagnosis of cancer. Data collection and analysis Two review authors independently assessed trial quality and extracted data. Meta-analyses were performed on different drug classes using continuous variable data. Main results Forty-five trials met the inclusion criteria. Only 27 of these trials involving 6746 participants were judged to have used a sufficiently robust measure of fatigue and thus were deemed suitable for detailed analysis. The drugs were analysed by class (psychostimulants; haemopoetic growth factors; antidepressants and progestational steroids). Methylphenidate showed a small but significant improvement in fatigue over placebo (Z = 2.40; P = 0.02). Erythropoietin showed a small but significant improvement in fatigue (for anaemic patients receiving chemotherapy) compared to placebo (Z = 2.67; P = 0.008). Darbopoietin also demonstrated a smaller but significant improvement in fatigue over placebo (Z = 1.96; P = 0.05). Paroxetine and progestational steroids demonstrated no superiority over placebo in treating CRF. There was a very high degree of statistical and clinical heterogeneity in the trials and the reasons for this are discussed. It was not possible to determine optimum doses as a result of this review. Authors' conclusions Erythropoietin and darbopoetin (for anaemic patients on chemotherapy) and psychostimulant trials provide evidence for improvement in CRF at a clinically meaningful level. There are no data to support the use of paroxetine or progestational steroids for the treatment of CRF. The obvious candidate drug for use in a large scale RCT is methylphenidate to confirm the preliminary results from this review.