The purpose of our study was to define a method and mechanism for overcoming the resistance of clinically relevant KRAS-mutant/LKB1-deficient NSCLC cells to the BET-bromodomain inhibitor JQ1.

LKB1 (Serine/threonine kinase 11) is mutated with loss of function in conjunction with mutated KRAS in 7-10% of NSCLC. Importantly, KRAS-mutant/LKB1-deficiency is associated with tumor aggressiveness and poor survival in human patients as well as in genetically engineered mouse models. Indeed, although the BET bromodomain inhibitor JQ1 dramatically reduces tumor volume in KRAS mutant mice, it has little effect in KRAS-mutant/LKB1-deficient mice. BET bromodomain proteins are chromatin readers that facilitate binding and activity of transcription factors such as the oncogene MYC. As such, JQ1 inhibits the transcriptional program of MYC and decreases MYC in mice.

To overcome JQ1 resistance in KRAS-mutant/LKB1-deficient NSCLC, we proposed to test JQ1 combination with chemotherapy, as such a combination would be a necessary arm of a clinical trial testing a BET bromodomain inhibitor. Here, we present data showing that gemcitabine permits and potentiates JQ1-mediated apoptosis of multiple KRAS-mutant/LKB1-deficient human NSCLC cell lines above either monotherapy. Increased apoptosis was observed by AnnexinV and TUNEL flow cytometry, PARP cleavage, and Comet assays.

Mechanistically, the drug combination had minimal effects on DNA damage and repair molecules in the KRAS-mutant/LKB1 deficient NSCLC cell line A549. Interestingly, however, JQ1 and combination with gemcitabine had profound effects on apoptotic molecules. JQ1 alone showed a robust increase in pro-apoptotic BimEL, BimL, and BimS that was balanced by a large increase in anti-apoptotic Mcl-1 24 hrs post-treatment. Similarly, MYC knockdown also increased Mcl-1, suggesting that JQ1 upregulation of Mcl-1 was at least partly dependent on JQ1 suppression of MYC. On the other hand, JQ1 combination with gemcitabine showed an increase in Bim that was not associated with increased Mcl-1.

Given these findings, we reasoned that KRAS-mutant/LKB1-deficient cells were resistant to JQ1-induced apoptosis executed by Bim because of protection provided by increased Mcl-1. In support of this hypothesis, Bim knockdown prevented JQ1-induced PARP cleavage in JQ1-sensitive H441 cells. Furthermore, Mcl-1 knockdown cells showed increased Bim, decreased Bcl-2, and increased PARP cleavage 48 hrs after JQ1 treatment in otherwise JQ1-resistant A549 cells.

Collectively, KRAS-mutant/LKB1-deficient cells are resistant to JQ1-induced apoptosis because of a compensatory increase in Mcl-1. This resistance can be overcome by combinatorial treatment with gemcitabine or direct Mcl-1 suppression, restoring apoptosis. Therefore, combinatorial treatment of a BET bromodomain inhibitor with gemcitabine or a Mcl-1 inhibitor may represent a potential novel strategy for treating the clinically relevant KRAS-mutant/LKB1-deficient subtype of NSCLC.

Citation Format: Michael Kahle, Margaret Soucheray, Jeffrey Becker, Eiki Kikuchi, Inés Pulido, Esra Akbay, Camilla Christensen, Wei Qiu, Fatima Al-Shahrour, Neil Johnson, Julián Carretero, Kwok-Kin Wong, Takeshi Shimamura. Overcoming KRAS/LKB1 mutant NSCLC resistance to BET bromodomain inhibitors with gemcitabine or Mcl-1 inhibition. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr C75.