Background: Insulin-like growth factors (IGF's) play a crucial role in\ncontrolling cancer cell proliferation, differentiation and apoptosis.\nExercise has been postulated as an effective intervention in improving\ncancer-related outcomes and survival, although its effects on IGF's are\nnot well understood. This meta-analysis aimed to determine the effects\nof exercise in modulating IGF's system in breast cancer survivors.\nMethods: Databases of PuMed, EMBASE, Cochrane Central Register of\nControlled Trials, EMBASE, ClinicalTrials. gov, SPORTDiscus, LILACS and\nScopus were systematically searched up to November 2014. Effect\nestimates were calculated through a random-effects model of\nmeta-analysis according to the DerSimonian and Laird method.\nHeterogeneity was evaluated with the I-2 test. Risk of bias and\nmethodological quality were evaluated using the PEDro score.\nResults: Five randomized controlled trials (n = 235) were included. Most\nwomen were post-menopausal. High-quality and low risk of bias were found\n(mean PEDro score = 6.2 +/- 1). Exercise resulted in significant\nimprovements on IGF-I, IGF-II, IGFBP-I, IGFBP-3, Insulin and Insulin\nresistance (P <0.05). Non-significant differences were found for\nGlucose. Aerobic exercise improved IGF-I, IGFBP-3 and Insulin. No\nevidence of publication bias was detected by Egger's test (p = 0.12).\nConclusions: Exercise improved IGF's in breast cancer survivors. These\nfindings provide novel insight regarding the molecular effects of\nexercise on tumoral microenvironment, apoptosis and survival in breast\ncancer survivors.