Breast cancer (BC) survivors have an increased risk of developing second primary cancer (SPC). The aim of this study was to detect and compare SPC predictors linked to the host, the first BC and its treatment. Two hundred and seventeen patients with a nonbreast SPC and 465 matched controls, nested in the cohort of BC patients diagnosed in a Spanish region between 1975 and 2003, were involved in a case-control study. The Tumour Registry database provided information about the host, BC and its treatment factors. Their contribution to the risk of developing SPC was measured by means of a conditional logistic regression. After controlling for differences between cases and controls at baseline, obesity [odds ratio (OR): 7.48; 95% confidence interval (CI): 1.25-44.88], smoking (OR: 3.16; 95% CI: 1.23-8.15), high blood pressure (OR: 1.68; 95% CI: 1.04-2.71) and having first-degree relatives suffering from cancer (OR: 1.69; 95% CI: 1.05-2.72) were the best SPC predictors. The risk of SPC increases by 1% per month of survival from BC (OR: 1.01; 95% CI: 1.007-1.012), while having metastases (OR: 0.23; 95% CI: 0.14-0.37) and being premenopausal at diagnosis of the BC (OR: 0.44; 95% CI: 0.247-0.792) diminish the risk, probably decreasing survival. The treatments were the regression model's worst predictors. Controlling modifiable factors linked to lifestyle such as obesity and smoking is essential to prevent SPC in survivors of BC. Health education to remove persistent risk factors should be included in the treatment protocol of BC patients, because they are important predictors of SPC.