X. Zhao, D. E. Mager; University at Buffalo, Buffalo, NY

BACKGROUND: Non-Hodgkin’s lymphoma (NHL) represents a heterogeneous B-cell neoplasm and the most common hematological cancer in adults. A diverse range of oncogenic mechanisms exists in lymphomagenesis creating challenges for developing NHL therapies. Discrete dynamic modeling is an excellent tool to analyze large regulatory networks and enhance understanding of complex biological systems. This study aimed to test the feasibility of using a network-based systems pharmacology analysis to identify intervention strategies based on molecular dysregulation in NHL.
METHODS: A Boolean model of B-NHL was constructed that incorporates B-cell receptor signaling, toll-like receptor and cytokine receptor pathways, intrinsic and extrinsic apoptosis, cell cycle arrest and DNA damage. In order to increase model predictability, we have been continuously updating the nodes and edges in the network based on most recent publications. Network visualization and centrality measures were performed in yEd graph editor. The network was further implemented into CellNetAnalyzer (CNA) for dynamic simulations. Logical steady states (LSS) and minimal intervention sets (MIS) were assessed.
RESULTS: The final B-NHL regulatory model contains 102 nodes and 180 edges. Common recurrent genetic alterations were considered by fixing nodes to either an “activated” or “inhibited” state. Centrality measures identified IKK/NFĸB, PI3K/AKT, p53/p21, Lyn/Syk/Btk and c-Myc/Bcl-6 as critical network hubs. LSS also predicted CD79B mutation as a proliferative marker in B-NHL. Based on MIS analysis, several combination interventions, including inhibitors of mTOR and Bcl-2, were suggested for further experimental evaluation.
CONCLUSION: A network-based systems pharmacology approach can be used to query key pharmacological targets in B-NHL and might provide a rational approach to design novel targeted combination therapies.