In the decade since the BRCA1 and BRCA2 genes were cloned, much has been learned about the function of these two major causes of familial breast cancer. BRCA2 has been shown to play a direct role in the repair of DNA by homologous recombination, by interacting with the Rad51 protein and facilitating the formation of Rad51 aggregates at the site of DNA damage. It likely plays a similar role when double strand breaks are created in the course of normal DNA replication; the absence of BRCA2 results in chromosomal instability, which is likely secondary to the defect in DNA repair. In the absence of BRCA2, the cell is more dependent on residual repair via Rad52, which makes Rad52 a target for therapy in BRCA-deficient tumors. BRCA1 plays a role in sensing DNA damage and replication stress and mediating the signaling responses. Therefore, in addition to its role in mediating DNA repair by homologous recombination via BRCA2, it can also signal cell cycle checkpoints and mediate other transcriptional responses to DNA damage. We have argued that the mechanism of cancer susceptibility from BRCA1 or BRCA2 deficiency is mediated via the defect in homologous recombination, since it is the main feature they share in common. We and others have recently demonstrated that the defect in homologous recombination changes the drug sensitivity profile, rendering the BRCA-deficient breast cancers sensitive to MitomycinC, cisplatin, etoposide and other drugs that produce complex double-stranded lesions in DNA. Furthermore, they show resistance to taxanes and navelbine. Fanconi anemia defective cells also show sensitivity to the same class of drugs, although their defect in homologous recombination in response to strand breaks appears less marked than in BRCA-deficient cells. However, Fanconi anemia cells also show chromosomal fragility, and appear to have defects in maintenance of the replication fork. Therefore, knowledge of whether this specific DNA repair pathway of homologous recombination is defective in breast cancer cells would be valuable information in planning optimized individual therapy. We have developed techniques to measure the functional integrity of homologous recombination in human breast cancers. Core biopsy samples are obtained and immediately irradiated ex vivo, allowing 3-4 hours for the appearance of Rad51, BRCA1 and FancD2 foci. Thin sections are obtained, permeabilized and stained by immunofluorescent techniques. We have identified tumors with defects in the ability to form Rad51 and BRCA1 foci, where there is no known genetic predisposition, implying that this BRCA-dependent repair pathway may be inactivated in sporadic as well as familial breast cancers. Thus, functional assays of homologous recombination could become a useful technique to determine phenotype of human breast cancer, which in turn will influence the choice of therapy.