Induction of Metabolic Impairment In Prolonged Early G1 Arrest Induced by CDK4/CDK6 Inhibition Sensitizes Myeloma Cells for Proteasome Inhibitor Killing During Subsequent S Phase Synchronization

Abstract 2989

Sequential drug combination is a rational approach to maximize tumor killing and minimize side effects in cancer therapy. However, this is rarely achieved because the mechanism of drug action is often incompletely understood and the cell cycle specificity of individual drugs unknown. Dysregulation of cyclin-dependent kinase (CDK)4 and CDK6 is common in human cancer and precedes unrestrained proliferation of tumor cells in multiple myeloma (MM) patients, especially during refractory relapse. This highlights the critical importance of targeting CDK4/CDK6 in MM.

We have now developed, for the first time, a novel therapeutic strategy to selectively inhibit CDK4/CDK6 in sequential combination with clinically relevant cytotoxic drugs for maximal tumor killing at reduced doses in MM. CDK4 and CDK6 promote reentry and progression of the cell cycle through G1. PD 0332991, the only known CDK4/CDK6-specific inhibitor, is potent, reversible and bioavailable. We showed that inhibition of CDK4/CDK6 with PD 0332991 induces early G1 arrest and upon release from the G1 block, synchronous progression to S phase and G2/M with exceptional precision and efficiency in MM cells in vitro and in animal models. This provides a unique means to determine the cell cycle targeting specificity of individual compounds for optimal combination. Simultaneous analysis of BrdU pulse-labeling (30 minutes) and DNA content per cell reveals that bortezomib, a reversible proteasome inhibitor; carfilzomib (PR-171), an irreversible selective inhibitor of the proteasom; and its oral analog ONX-0912 (PR-047) all preferentially target MM cells synchronized into S phase over those arrested in G1, but not cells in G2/M. On this basis, killing of myeloma cells by proteasome inhibitors is markedly enhanced in prolonged G1 arrest induced by PD 0332991 and further augmented during synchronous entry into and progression through S phase upon release from the G1 block, in vitro and in vivo in the native bone marrow niches.

Induction of early G1 arrest by PD 0332991 requires Rb, the substrate of CDK4 and CDK6, but not p53. Importantly, the increase in carfilzomib, ONX-0912 or bortezomib mediated killing after S phase synchronization significantly surpasses the enrichment of S phase cells. It is in fact proportional to the time of prior G1 arrest. Kinetics analyses of global gene expression patterns, specific RNA and protein levels and functional shRNA interference show that prolonging early G1 arrest leads to time-dependent uncoupling of gene expression from the cell cycle. PD 0332991 withdraw rapidly restores the CDK4 and CDK6 catalytic activity and scheduled expression of cell cycle genes, hence synchronous progression to S phase and mitosis. This includes upregulation of cyclin A synthesis and Skp2 mediated ubiquitin-proteasome degradation of p27 for S phase entry, mini chromosome maintenance(MCM)s and thymidine kinase for DNA replication, and genes critical for G2/M checkpoint control and mitosis. However, it fails to fully reverse the metabolic impairment (altered glucose, nucleotide and ATP metabolism) induced in prolonged early G1 arrest. This culminates in the loss of IRF-4 required for myeloma survival and selective gain of pro-apoptotic Bim function in G1 arrest and Noxa in S phase in synergy with carfilzomib and bortezomib, which lowers the threshold for activation of the intrinsic apoptosis pathway.

Selective inhibition of CDK4/CDK6 in sequential combination therapy thus not only halts tumor cell proliferation but also potently induces synergistic tumor killing. This sequential combination therapy has been implemented in a multi-center phase 1/2 clinical trial targeting CDK4/6 with PD 0332991 in combination with bortezomib and dexamethasone in relapsed refractory MM. Phase 1 data indicate that PD 0332991 is well tolerated, and directly and completely inhibits CDK4/CDK6 and the cell cycle in tumor cells in MM patients with promising clinical efficacy. Evidence from phase 2 trials of carfilzomib indicates that it is also well tolerated. The peripheral neuropathy commonly observed with bortezomib appears to be less severe and less frequent. Selective combination with carfilzomib or the oral agent ONX-0912 thus represents a promising alternative to refine targeting CDK4/6 in sequential combination therapy for multiple myeloma and potentially other cancers.