Multiparity decreases the risk of breast cancer in white women, whereas it is a risk factor in black women o50 years. Early-onset breast cancer (o50 years) has been associated with high insulin-like growth factor-1 (IGF-1) levels. Absence of the common IGF1 19 cytosine-adenine (CA)-repeat allele (IGF1-19/-19) inverts the effect of several non-genetic factors on breast cancer risk but the interaction between IGF1-19/-19 and multiparity on breast cancer risk is unknown. As IGF1-19/-19, multiparity and early-onset breast cancer are more common in black than in white women, we aimed to study whether multiparity combined with IGF1-19/-19 increases the risk of early-onset breast cancer. Four hundred and three breast cancer patients diagnosed in Lund, Sweden, at age 25 -99 years were genotyped for the IGF1 CA-repeat length using fragment analysis. Overall, 12.9% carried the IGF1-19/-19 genotype. There was a highly significant interaction between multiparity and IGF1-19/-19 on age at breast cancer diagnosis (P ¼ 0.007). Among IGF1-19/-19 patients, multiparity was associated with a 9.2 year earlier age at diagnosis compared with uniparity or nulliparity (P ¼ 0. Breast cancer is the most common malignancy in women worldwide (Parkin et al, 2001(Parkin et al, , 2002. Approximately 10% of women living in Sweden will develop the disease during their lifetime. The median age at breast cancer diagnosis in women in Sweden is 60 years; approximately 4% of women are 40 years or younger at diagnosis (data from the Swedish Cancer Registry). The prognosis is often poorer in younger than in older women (Sidoni et al, 2003) in terms of overall survival and relapse (Han et al, 2004), partly because young women do not benefit from breast cancer screening and early-onset breast cancers are usually detected clinically at a more advanced stage. In Lund, Sweden, mammography screening is initiated at age 45 years and is performed every 18 months until the age of 74 years. As of today, no tests are routinely performed to identify the subgroup of women who do not belong to high-risk breast cancer families (eg BRCA1 families) but who would nevertheless benefit from breast cancer screening at an earlier age owing to an increased risk for early-onset breast cancer.Breast cancer risk is determined by genetic factors and nongenetic factors, for example, reproductive factors (Beral and Reeves, 1993;Kelsey et al, 1993). An increased age at first full-term pregnancy and low parity have been associated with an increased risk of developing breast cancer in the developed world (Kelsey et al, 1993). In the short term, a first full-term pregnancy increases the risk of breast cancer (Beral and Reeves, 1993). Each subsequent live-birth reportedly decreases the risk of breast cancer by 30% and after three full-term pregnancies, the short-term risk of developing breast cancer is similar to the risk of nulliparous women (McCredie et al, 1998). In the long-term, a full-term pregnancy confers a decreased risk of developing breast cancer (McCredie et al, 1998) and ...