Editors' ChoiceCancer

PARP inhibitors need an extra STING for therapeutic efficacy

See allHide authors and affiliations

Science Translational Medicine  29 May 2019:
Vol. 11, Issue 494, eaax9557
DOI: 10.1126/scitranslmed.aax9557

Abstract

PARP inhibitors require the recruitment of CD8+ T cells through STING-cGAS signaling to maximize tumor regression.

Cancer cells with homologous recombination (HR) deficiency are selectively killed by poly(ADP) ribose polymerase (PARP) inhibitors in vitro and in vivo, which has led to the approval of PARP inhibitors in HR-deficient ovarian and breast cancers. In addition to their direct cytotoxic effects, it is now recognized that PARP inhibitors and other genotoxic agents lead to immune modulatory changes in treated cells. Pantelidou and colleagues extend these findings to describe the mechanism and in vivo relevance of immune activation after PARP inhibitor treatment. Using a genetically engineered mouse model of BRCA-mutant mammary carcinoma, they observed that CD8+ T cell depletion markedly reduces the therapeutic efficacy of PARP inhibition. Interestingly, PARP inhibitor treatment selectively lead to Type 1 interferon production through activation of STING-cGAS in HR-deficient, but not in HR-proficient, cells. CRISPR-Cas9–mediated knockout of STING also blunted the efficacy of PARP inhibitors in shrinking tumors in this model system. The authors conclude that CD8+ T cell recruitment through the induction of the STING-cGAS system is a critical determinant of PARP inhibitor activity in vivo.

The findings in this paper are largely based on a murine model of BRCA-mutant breast cancer and need to be validated in other HR-deficient cancer models and human samples. If validated, these findings have important implications for both clinical trial design and preclinical research. If PARP inhibitors promote T cell recruitment in HR-deficient cancers, they could be combined in this setting with agents that impede T cell checkpoints. Conversely, in HR-proficient tumors, PARP inhibition is not expected to generate Type 1 interferon responses, suggesting that empiric combinations of PARP inhibitors with immune checkpoint inhibitors are unlikely to synergize. For drug discovery, the marked difference in efficacy of PARP inhibitors in immunocompetent and immunodeficient mice raises a provocative question as to whether potentially useful anticancer drugs are missed through in vitro screens and assays that do not account for the immune system.

Highlighted Article

View Abstract

Navigate This Article