Rad51 protein, involved in homologous recombination, is overexpressed in a variety of tumors, and its expression is correlated with a poor prognosis. Here we propose to exploit the overexpression of Rad51 in cancer cells to design a Rad51 promoter-based anticancer therapy. On average, Rad51 mRNA and protein levels are increased in cancer cells four- and sixfold, respectively. Serendipitously, we discovered that when the Rad51 ORF is replaced with another ORF, the difference in promoter activity between normal and cancer cells increases to an average of 840-fold with a maximum difference of 12,500-fold. This dramatic difference in activity has high therapeutic potential. We demonstrate that the fusion of Rad51 promoter to diphtheria toxin A (DTA) gene kills a variety of cancer cell types, including breast cancer, fibrosarcoma, and cervical cancer cells, with minimal effect on normal breast epithelial cells and normal fibroblasts. Our results suggest that therapies based on the Rad51 promoter will be highly tumor specific and open new avenues for targeting a broad range of cancers.