To assess the association between coffee intake and breast cancer risk using data from studies published during the past 15 years. Articles published between January 2005 and May 2020 were collected from PubMed, Embase, and Web of Science databases, and the Cochrane library. Coffee consumption was set as the exposure factor of breast cancer risk, and relative risk (RR) was set as the assessment indicator. Random-effects or fixed-effects models were used for the meta-analysis, depending on the level of heterogeneity. The linear and non-linear dose-response relationship was assessed using the generalized least-squares method and restricted cubic spline model. Articles were evaluated by sensitivity analysis, and Begg's and Egger's tests and funnel plots were used to evaluate publication bias. The process of meta-analysis referred to Meta-analyses of Observational Studies in Epidemiology (MOOSE). We retrieved 26 relevant studies in the dose-response analysis and meta-analysis. A negative correlation was found between coffee consumption and breast cancer risk (RR: 0.95; 95% CI: 0.92-0.99). A linear and marginal dose-response relationship was found for six case-control studies (RR: 0.98; 95% CI: 0.95-1.00) and 10 cohort studies (RR: 0.98; 95% CI: 0.97-1.00). Subgroup analysis showed that the relationship between coffee consumption and breast cancer risk was moderated by menstrual status and geographic region. Overall, the meta-analysis found a negative correlation between coffee intake and breast cancer risk, especially in postmenopausal and European women.