Polymorphisms in the mitochondrial genome are hypothesized to be associated with risk of various diseases, including cancer. However, there has been conflicting evidence for associations between a common polymorphism in the mitochondrial genome (A10398G, G10398A in some prior reports) and breast cancer risk. Reactive oxygen species, a by-product of mitochondrial energy production, can lead to oxidative stress and DNA damage in both the mitochondria and their cells. Alcohol consumption, which may also lead to oxidative stress, is associated with breast cancer risk. Therefore, we hypothesized that polymorphisms in the mitochondrial genome interact with alcohol consumption to alter breast cancer risk. We genotyped the A10398G polymorphism in a case-control study nested within the Nurses' Health Study (NHS, 1,561 cases, 2,209 controls). We observed an interaction between alcohol consumption (yes/no) and A10398G on breast cancer risk (p-int = 0.03). The risk associated with alcohol consumption was limited to carriers of the 10398G allele (Odds Ratio 1.52, 95% Confidence Interval 1.10-2.08 comparing drinkers to non-drinkers). However, we were unable to replicate these findings in the Women's Health Study (WHS, 678 cases, 669 controls), although the power to detect this interaction in the WHS was low (power = 0.57). Further examination of this interaction, such as sufficiently powered epidemiological studies of cancer risk or associations with biomarkers of oxidative stress, may provide further evidence for GxE interactions between the A10398G mitochondrial polymorphism and alcohol consumption on breast cancer risk.