Systems pharmacology modeling of bortezomib in cellular and xenograft myeloma systems
Chudasama, Vaishali L.
MetadataShow full item record
Multiple myeloma (MM) is a plasma cell (terminally differentiated B cell) malignancy that is mostly characterized by monoclonal expansion of these cells in the bone marrow. Despite the availability of multiple treatment options (e.g., stem cell transplantation and combinatorial chemotherapy), a majority of patients relapse and a significant fraction of patients are non-responders. The design of current combination regimens is empirical in nature, and these regimens are often accompanied by low treatment adherence due to severity of side effects and/or lack of treatment effectiveness. Bortezomib, a potent and reversible proteasome inhibitor, is approved for the treatment of newly diagnosed and relapsed/refractory MM and targets multi-subunit protein complexes involved in cell survival via regulation of protein turnover in cells. Mechanisms underlying bortezomib-mediated cell cytotoxicity are poorly understood, and further research is needed to map these mechanisms and characterize the dynamic behavior of intracellular proteins involved in the regulation of cell cytotoxicity. The major goal of this thesis is to develop a mechanism-based cellular pharmacodynamic model of bortezomib effects in MM cell and xenograft systems that can enhance understanding of bortezomib mechanisms of action, identify critical factors regulating bortezomib signal transduction, and predict bortezomib responses in combination with curcumin. Evaluating mechanisms of drug action is critical for establishing exposure-response relationships with the potential to guide precision medicine. In Chapter 2, a Boolean network model is developed to better understand and predict bortezomib mechanisms of action. The final Boolean network model was established using pathway specific probes (IKKi, JAKi, and IκBi) and qualified using experimental western blot and cell proliferation studies. The final network was used to simulate bortezomib responses, and simulated results were in good agreement with experimental data. Model simulations suggested that stress accumulation upon proteasome inhibition is a major pathway leading to myeloma cell death, which contradicts another major hypothesis that states inhibition of the NFκB pathway is the major mechanism leading to myeloma cell death. Systems-level modeling may facilitate the rational design of single-agent and combinatorial dosing regimens, approaches to overcome drug resistance, and regimens that avoid sub-optimal concentrations. In Chapter 3, our Boolean network model is reduced and essential factors that regulate bortezomib signal transduction were identified (e.g., pNFκB, BclxL, and cleaved PARP). These factors were integrated into a cellular pharmacodynamic model of multiple biomarkers following continuous bortezomib exposure (20 nM) in U266 cells. The reduced dynamic model was successfully extended to characterize the effects of bortezomib on the time-course of tumor volume in U266 xenografts. Curcumin is a naturally occurring polyphenolic plant (Curcuma longa) product. It is being evaluated in combination with bortezomib for the treatment of MM. In Chapter 4, our cellular pharmacodynamic model developed for bortezomib was modified to incorporate curcumin-mediated alteration in U266 cell signaling. Responses to curcumin were successfully characterized after continuous exposure to curcumin (50 μM). This cellular pharmacodynamic model for curcumin was also successfully extended to describe U266 xenograft tumor volume profiles and predict curcumin exposure-response relationships at 48 h. The combination of two or more drugs has become common practice to decrease adverse events and the development of resistance to monotherapy for different disease types including cancer. In vivo studies of curcumin in combination with bortezomib have shown to potentiate bortezomib anti-tumor effects. Evaluating this drug combination using a PK-PD approach could help in quantifying the time-course and extent of this interaction that could potentially facilitate the rational design of combination regimens. In Chapter 5, a unifying pharmacodynamic model was developed to reconcile the in vitro and in vivo drug-drug interactions with bortezomib and curcumin. The combination model successfully predicted in vitro antagonistic and in vivo synergistic effects for the bortezomib-curcumin combination. The model may prove to be useful for designing in vivo combination studies and predict outcomes to bortezomib-based combinations. This dissertation attempts to integrate network-based and systems pharmacology approaches to improve understanding of bortezomib mechanisms of action and define PK-PD relationships for bortezomib and its use in combination regimens. Whereas algorithms and approaches are available for reducing and calibrating mechanistic biochemical models, new methods are needed for developing multi-scale models of drug action for cases in which such biochemical models are not yet available. We show how network-based analysis, a major component of systems biology, and in vitro cell response data may be used in an integrated manner to guide the development of multi-scale pharmacodynamic models. Although further research is needed to refine this approach, it is flexible and multi-purpose. Network models may be readily extended to include additional signaling pathways and multiple cell types to assess novel drug targets and combinations, disease progression, drug resistance, and multiple potential sources of inter-individual variability in therapeutic responses. (Abstract shortened by UMI.)