Antitherapeutic antibody-mediated hepatotoxicity of recombinant human Apo2L/TRAIL in the cynomolgus monkey.

Apo2L/TRAIL is a member of the tumor necrosis factor superfamily and an important inducer of apoptosis. Recombinant human (rhu) Apo2L/TRAIL has been attractive as a potential cancer therapeutic because many types of tumor cells are sensitive to its apoptosis-inducing effects. Nonclinical toxicology studies were conducted to evaluate the safety of rhuApo2L/TRAIL for possible use in humans. The cynomolgus monkey was chosen for this safety assessment based on high protein sequence homology between human and cynomolgus Apo2L/TRAIL and comparable expression of their receptors. Although hepatotoxicity was observed in repeat-dose monkey studies with rhuApo2L/TRAIL, all animals that displayed hepatotoxicity had developed antitherapeutic antibodies (ATAs). The cynomolgus ATAs augmented the cytotoxicity of rhuApo2L/TRAIL but not of its cynomolgus counterpart. Of note, human and cynomolgus Apo2L/TRAIL differ by four amino acids, three of which are surface-exposed. In vivo studies comparing human and cynomolgus Apo2L/TRAIL supported the conclusion that these distinct amino acids served as epitopes for cross-species ATAs, capable of crosslinking rhuApo2L/TRAIL and thus triggering hepatocyte apoptosis. We describe a hapten-independent mechanism of immune-mediated, drug-related hepatotoxicity - in this case - associated with the administration of a human recombinant protein in monkeys. The elucidation of this mechanism enabled successful transition of rhuApo2L/TRAIL into human clinical trials.

ODmixed: a tool to obtain optimal designs for heterogeneous longitudinal studies with dropout.

ODMixed is a computer program to obtain optimal designs for linear mixed models of longitudinal studies. These designs account for heterogeneous correlated errors and for data with dropout. Designs are compared by using relative efficiencies, e.g., between a D-optimal design for homogeneous data and another for heterogeneous data or between a D-optimal design for complete data against another that optimizes designs when data is missing at random. Two examples are worked out to illustrate how researchers could use this computer program to profit of optimal design theory at the planning stage of longitudinal studies.