Poly(ADP-ribose) polymerase inhibition (PARPi) is a promising new therapeutic approach for the treatment of cancers that show homologous recombination deficiency (HRD). Despite the success of PARPi in targeting HRD in tumors that lack BRCA1 or BRCA2 function, drug resistance poses a major obstacle to PARPi efficacy. Several resistance mechanisms have been identified in cultured cells, but testing their relevance for in vivo resistance is cumbersome. As a novel approach to investigate these, we describe here the use of 3D cancer organoids derived from genetically engineered mouse models (GEMMs) for BRCA1/2-deficient cancers. Unlike conventional cell lines or mammospheres, organoid cultures can be efficiently derived and rapidly expanded in vitro. Orthotopically transplanted organoids give rise to mammary tumors that recapitulate the epithelial morphology and preserve the drug response of the original tumor. Importantly, GEMM tumor-derived organoids can be easily genetically modified providing a powerful tool for genetic studies.