Antigenic modulation limits the efficacy of anti-CD20 antibodies: implications for antibody selection.

Rituximab, a monoclonal antibody that targets CD20 on B cells, is now central to the treatment of a variety of malignant and autoimmune disorders. Despite this success, a substantial proportion of B-cell lymphomas are unresponsive or develop resistance, hence more potent anti-CD20 monoclonal antibodies (mAbs) are continuously being sought. Here we demonstrate that type II (tositumomab-like) anti-CD20 mAbs are 5 times more potent than type I (rituximab-like) reagents in depleting human CD20 Tg B cells, despite both operating exclusively via activatory Fcgamma receptor-expressing macrophages. Much of this disparity in performance is attributable to type I mAb-mediated internalization of CD20 by B cells, leading to reduced macrophage recruitment and the degradation of CD20/mAb complexes, shortening mAb half-life. Importantly, human B cells from healthy donors and most cases of chronic lymphatic leukemia and mantle cell lymphoma, showed rapid CD20 internalization that paralleled that seen in the Tg mouse B cells, whereas most follicular lymphoma and diffuse large B-cell lymphoma cells were far more resistant to CD20 loss. We postulate that differences in CD20 modulation may play a central role in determining the relative efficacy of rituximab in treating these diseases and strengthen the case for focusing on type II anti-CD20 mAb in the clinic.

Interplay between function and structure in complex networks.

We show that abrupt structural transitions can arise in functionally optimal networks, driven by small changes in the level of transport congestion. Our results offer an explanation as to why so many diverse species of network structure arise in nature (e.g., fungal systems) under essentially the same environmental conditions. Our findings are based on an exactly solvable model system which mimics a variety of biological and social networks. We then extend our analysis by introducing a renormalization scheme involving cost motifs, to describe analytically the average shortest path across multiple-ring-and-hub networks. As a consequence, we uncover a "skin effect" whereby the structure of the inner multi-ring core can cease to play any role in terms of determining the average shortest path across the network.

Effect of congestion costs on shortest paths through complex networks.

We analyze analytically the effect of congestion costs within a physically relevant, yet exactly solvable, network model featuring central hubs. These costs lead to a competition between centralized and decentralized transport pathways. In stark contrast to conventional no-cost networks, there now exists an optimal number of connections to the central hub in order to minimize the shortest path. Our results shed light on an open problem in biology, informatics, and sociology, concerning the extent to which decentralized versus centralized design benefits real-world complex networks.