P453: CDK6 DEGRADATION IS IMPEDED BY P16INK4A AND P18INK4C IN AML

Abstract

Background: Acute myeloid leukemia (AML) is a highly aggressive disease that comprises a heterogeneous group of genetically distinct subtypes. AML is initiated by leukemic cells, which are able to self-renew and give rise to malignant myeloid blasts. A key regulator and prognostic biomarker for AML is the cell cycle regulator Cyclin-dependent kinase 6 (CDK6) which represents a novel therapeutic target for the treatment of certain subtypes of AML. CDK4/6 kinase inhibitors are extensively studied in several cancer types but the beneficial effects may be limited by primary and secondary resistance mechanisms. Pharmacological CDK6 degraders, which eliminate kinase -dependent and -independent effects, represent an alternative therapeutic option. The exact mechanism and efficacy of CDK6 degraders in AML subtypes remain unknown. The involvement of p16INK4A in Palbociclib resistance and the role of INK4 proteins in AML disease progression require a systematic investigation of INK4 proteins in the context of CDK6 degradation in distinct AML subtypes. Aims: We aim to elucidate CDK6 degrader efficacy in two CDK6-dependent AML subtypes harboring either MLL-AF9 or AML1-ETO. We hypothesize an involvement of INK4 proteins on CDK6 degrader efficacy. Understanding the exact mechanisms of the CDK6 degrader will help to determine AML subtypes that would benefit from CDK6-targeted therapy. Methods: We took advantage of a novel hematopoietic progenitor cell model (HPCLSKs) to study CDK6 degrader efficacy in a genetically defined setting of AML subtypes. We analyzed INK4 RNA expression of human AML patients and the HPCLSK cells using RNA-Seq and Microarray data sets. We analyzed cell proliferation and cell cycle changes using flow cytometry. CDK6 Co-immunoprecipitation followed by Mass Spectrometry was performed to screen for CDK6 interaction partners that are enriched upon CDK6 Degrader treatment. These findings were validated via immunoblotting. Human AML cell lines were used to further validate the results in a different in vitro system and to translate them to human settings. Results: We show that efficacy of the CDK6 specific protein degrader varies among AML subtypes and depends on low expression of the INK4 proteins p16INK4A and p18INK4C. INK4 protein levels are significantly elevated in MLL-AF9+ compared to AML1-ETO+ cells, contributing to the different CDK6 degradation efficacy. We demonstrate that CDK6 complexes containing p16INK4A or p18INK4C are protected from pharmacological degradation and that INK4 levels define the proliferative response to CDK6 degradation. These findings define INK4 proteins as predictive marker for CDK6 degradation – targeted therapies in AML. Summary/Conclusion: We here identify p16INK4A and p18INK4C as dominant CDK6 binding partners that counteract pharmacological protein degradation in AML. Our data highlight the need for novel CDK6 specific therapies targeting CDK6 sites not competing with INK4 or CIP/KIP binding to overcome therapeutic limitations.