Background: Physical activity and exercise has been identified as a curative intervention to improve physical and mental health, and even cancer prognosis, in breast cancer survivors. Though there have been many randomized clinical trials (RCTs) that attempted to prove the effect of exercise on cancer-related biomarkers, including fasting insulin level and inflammatory markers, in breast cancer survivors, the results has not been consistent. Thus, this systematic review and meta-analysis aimed to update and synthesize the biomarker outcomes from RCTs evaluating effects of structured exercise training in breast cancer survivors. Methods: MEDLINE, Cochrane-CENTRAL, EMBASE, and SPORTDiscus databases were searched until December 2013. We included RCTs that investigated whether exercise intervention can change cancer-related biomarkers in blood compared with a control group in breast cancer survivors. The primary outcome was fasting insulin level, and the secondary outcomes were fasting glucose level, insulin resistance, insulin-like growth factor 1 (IGF-1), insulin-like growth factor binding protein 3 (IGFBP-3), and interleukin 6 (IL-6). Pooled standard mean differences (SMD) in each outcome between intervention and control groups were calculated using a random-effects model. Results: Of 1,256 articles retrieved, 17 articles (674 participants, 14 RCTs) were included in the analysis. Overall, insulin levels significantly reduced after exercise intervention in the exercise group (SMD -0.80; 95% confidence interval [CI] -1.56 to -0.04; p=0.041) compared with the control participants. Fasting glucose level, insulin resistance, IGF-1, IGFBP-1, IGFBP-3, and IL-6 were not different between exercise and control groups after intervention. Conclusions: This meta-analysis proved that exercise can significantly reduce fasting insulin level in breast cancer survivors, which may positively affect breast cancer prognosis.