Methods 13 articles published before 31 December 2019 involving 1154 participants were selected according to PICO guidelines in the Cochrane Handbook. Relevant randomized controlled trials (RCTs) were included. Hedges d effect sizes were calculated and random effects models were used to estimate the pooled effects. I2 tests were applied to assess the heterogeneity. Moderating effects were tested by mixed model meta-regression analysis according to moderators derived from participant characteristics, features of Qigong exposure and research design. Study quality was judged using the Wayne Checklist. Results Qigong practice relieved cancer-related fatigue by a heterogeneous (I2 = 81.4%) standardized mean effect size 0.46 (95% CI, 0.15 to 0.78, z = 2.89, p = 0.0039). Reductions were larger in participants having elevated fatigue at baseline. Trials with blinded allocation or blinded assessment of participants had larger effects. Qigong had a significant effect on cancer-related fatigue when Qigong was compared with usual care or waitlist control (Hedges d = 0.66, 95% CI, 0.07 to 1.26, p < 0.001), but not when Qigong was compared with Western exercise (Hedges d = 0.46, 95%CI, −0.02 to 0.95, p = 0.06) or no treatment control (Hedges d = 0.10, 95%CI, −0.23 to 0.43, p = 0.60) in sub-analysis. Conclusion Qigong practice may have small-to-moderate efficacy for management of cancer-related fatigue, but the limited number of RCTs and methodological flaws in most of the trials make it premature to conclude clinical effectiveness. Further high-quality RCTs are needed to mitigate potential methodological bias.