PARP inhibitor MK-4827 is synthetic lethal for tumors with homologous recombination defects associated with ATM-deficiency, PTEN-deletion and microsatellite instability (MSI)

Abstract

Poly(ADP-ribose) polymerase (PARP) -1 and -2 enzymes play a critical role in the base excision DNA repair (BER) pathway. PARP inhibition in oncology has the potential to target pre-defined sub-populations with higher sensitivity and to widen the therapeutic index of chemotherapy and radiotherapy. Pre-clinical and clinical evidence demonstrated that PARP inhibitors are synthetic lethal for tumors with defects in the homologous recombination (HR) pathway, such as tumors with mutations in the BRCA1 or BRCA2 genes. MK-4827 is a potent and selective orally available PARP-1/2 inhibitor currently in Phase 1 development. In pre-clinical models, MK-4827 displays excellent monotherapy activity in a large panel of BRCA mutant cell lines (10 nM ≤ CC50 ≤ 90nM) with at least 10-fold selectivity over BRCA wild-type celllines (CC50 ≥ 1.5 uM). We have explored the potential of MK-4827 for the therapy of tumors with defects in the HR pathway in addition to mutation in the BRCA genes. In this study we have evaluated the activity of MK-4827 on tumor cell lines with defects in the homologous recombination pathway as a consequence of 1) inactivation of ATM gene in melanoma and mantle cell lymphoma cell lines; 2) PTEN gene deletion in prostate cancer cells; 3) MRE11 expression level defects associated with the Microsatellite Instability (MSI) phenotype in colorectal cancer cells (CRCs). We demonstrated that cancer cells with an inactive ATM pathway as a consequence of homozygous ATM gene inactivation are very sensitive to MK-4827 inhibition (CC50≤ 50nM). MK-4827 also displays strong activity in ATM-deficient xenograft models. Heterozygous ATM inactivation is not sufficient to confer sensitivity to MK-4827. Our studies also showed that prostate cancer cells with homozygous-deletion of PTEN gene are sensitive to MK-4827 inhibition (100nM ≤ CC50 ≤ 600nM) as measured by long-term cell growth and proliferation assays. Homozygous, but not heterozygous, PTEN-deletion is required for MK-4827 sensitivity in agreement with data available for sensitivity of BRCA- and ATM-deficient cells to MK-4827. Finally, we have screened a panel of CRC cells with known Microsatellite Instability and have observed that MSI+ cells are sensitive to MK-4827 in long term proliferation assays, in contrast to Microsatellite stable (MSS) CRC cells which are resistant. In agreement with previous reports, we observed that MSI+ CRC cells show low expression levels of MRE11, a key protein for homologous recombination. Conclusion: Our pre-clinical data demonstrate that MK-4827 is synthetic lethal for tumors with ATM-deficiency, PTEN-deletion or MSI+ instability and suggest that patients with tumors carrying these genetic defects might benefit from treatment with PARP inhibitors.