CONTEXT: Light therapy is a non-pharmacological therapy that is currently being studied in cancer-related symptoms and is certificated as a low-risk intervention by FDA. Cancer-related fatigue (CRF) is the most common symptom reported by cancer patients. OBJECTIVE: To examine the effectiveness of light therapy for CRF in cancer patients through a systematic review and meta-analysis. METHODS: We conducted a systematic review of four electronic databases targeted randomized clinical trials evaluating light therapy for CRF (CRD42020215446), from inception to May 2021. The primary outcome was changes of CRF scores; secondary outcomes included depression, sleep, and quality of life (QoL). We quantitatively pooled outcomes using meta-analysis with random-effects models and assessed methodological bias. RESULTS: We identified thirteen RCTs representing 551 cancer patients, encompassing breast (n = 5), ovarian or endometrial (n = 1), multiple myeloma (n = 1), lung (n = 1), or combined (n = 5) cancers. The comparison groups included dim light (n = 12) and waiting list (n = 1). Duration of intervention ranged from 1 to 12 weeks. Light intensities ranged from 417.9 to 12,000 lux. Light therapy was associated with a significant improvement in CRF (SMD = 0.45, P = 0.007), depression (SMD = -0.26, P = 0.03) and sleep difficulty (SMD = -2.46, P = 0.0006); a statistically non-significant trend was observed for QoL (SMD = 0.33, P = 0.09). Funnel plots for CRF suggest not significant publication bias. CONCLUSION: Light therapy could be a feasible and effective option for improving CRF in cancer patients. Larger sample, rigor trials design and a standard protocol of intervention are needed to draw more conclusive conclusions.