Exploring the Impact of Melatonin on Cancer-Related Fatigue: A Dose-Response Meta-Analysis With Grade Evidence Evaluation Publisher Pubmed



Etesamnia S ; Nasiraie Moghadam M ; Mostafazadeh P ; Jowshan M R ; Roudini K ; Miri S R ; Jahan Mihan A ; Haghighi A ; Ebrahimpour Koujan S
Source: BMC Cancer Published:2026

Abstract

Background: Cancer-related fatigue (CRF) is a prevalent and debilitating symptom among patients undergoing cancer treatment, yet its management remains a clinical challenge. Previous studies evaluating the role of melatonin in CRF have yielded inconsistent findings. This meta-analysis aimed to assess the overall effect of melatonin supplementation on fatigue and to explore potential dose–response relationships. Methods: A systematic literature search was performed in PubMed, Scopus, Web of Science, and Google Scholar up to 24 August 2025. Randomized controlled trials comparing melatonin with a placebo in adult cancer patients were included. SMDs with 95% CIs were calculated using a random-effects model. Nonlinear and linear dose–response analyses were conducted, and certainty of evidence was evaluated with the GRADE approach. Results: Twelve trials (18 effect sizes, n = 1,977) met the inclusion criteria. Pooled analysis showed a significant reduction in fatigue scores with melatonin supplementation (SMD: −0.45; 95% CI: −0.69, − 0.20; P < 0.001). Dose–response analysis revealed a significant inverse association for melatonin doses between 5 and 15 mg/day (P-nonlinearity = 0.03). Short-term interventions (≤ 4 weeks) demonstrated greater benefit, whereas longer durations showed no consistent effects. However, substantial heterogeneity (I² = 78%) and very low certainty of evidence limit the robustness of these findings. Conclusions: Melatonin supplementation may reduce fatigue in cancer patients, particularly at doses of 5–15 mg/day and with short-term use. Nonetheless, the certainty of evidence remains very low, highlighting the need for high-quality, large-scale RCTs with standardized methodologies to validate these results. © The Author(s) 2026.