Essential role of the cyclin-G associated kinase (GAK) in diffuse large B-cell lymphoma progression - Project Summary/Abstract. Diffuse large B-cell lymphoma (DLBCL) is the most frequently diagnosed lym- phoma and carries poor prognosis for the ~40% of patients not cured by frontline therapy. While emerging im- munotherapies provide new options for relapsed or refractory (rel/ref) patients, kinase inhibitors have found little role in DLBCL management despite transformational impact on other lymphomas. In this proposal, we de- fine the cyclin-G associated kinase (GAK) as a novel therapeutic target, discovered through phenotypic screening and machine-learning target deconvolution. We find GAK is a cell-cycle kinase critical for mitotic spindles. DLBCL tumors with reduced function of the tumor suppressor retinoblastoma (RB) in particular show critical dependence on GAK. Preliminary data also demonstrate in vivo efficacy of GAK inhibition against DLBCL xenografts, employing a pharmacologically unfavorable tool compound. To leverage this discovery to rapid evaluation in rel/ref patients, we found clinically available kinase inhibitors with strong anti-GAK activities suitable for repurposing. Several of these have stronger activity against GAK than their primary clinical targets and demonstrate potent killing of DLBCL tumors. We also reveal phosphoproteomics that implicate for the first time direct substrates by which GAK regulates mitosis. Our central hypothesis is that rapidly dividing DLBCL cells require GAK’s phosphorylation of specific mitotic regulators for successful nucleation and positioning of mitotic spindles. To test this, we pursue three specific aims. Aim 1 assesses microscopically the context and mechanisms GAK kinase regulation of mitosis. Confocal microscopy optimized on a high-throughput live-imag- ing platform seek to define specifically the role of GAK kinase activity in mitosis and the mechanistic basis for increased GAK dependence in RB-deficient tumors. Aim 2 defines biochemically and functionally GAK’s phos- phorylation substrates that regulate mitosis. Substrates to be evaluated biochemically and functionally include MTUS2, CENPF, EZRIN, and other candidates from our preliminary data. Aim 3 evaluates anti-lymphoma effi- cacy and hematologic toxicity of GAK inhibition in vivo. Inhibitors of other mitotic kinases like Aurora kinase A have been plagued clinically by hematologic toxicities. Here, using complementary murine lymphoma models, we pursue a hypothesis that GAK inhibition will be efficacious at reduced toxicity. Moreover, results are de- signed to directly inform development of investigator initiated trials (IITs) of repurposed compounds from this project co-led by a lymphoma physician-scientist. We expect to define GAK – never before studied as a can- cer therapeutic target – as a critical dependency for mitosis by DLBCL. We both investigate mechanisms and rigorously determine GAK’s suitability as a novel clinical target. Our project is innovative because we show for the first time GAK is a mitotic kinase and the experimentally generated phospho-substrates by which it carries out these activities. Our project is significant because unmet clinical need remains persistently high for rel/ref DLBCL patients, and RB deficiency is a common, high-risk, and currently untargetable biomarker in patients.
