Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL, Apo2L) has been shown to exhibit potent and specific apoptotic activity against tumor cells. Several TRAIL constructs have been tried in patients, and the molecule remains under active clinical investigation. Native and recombinant TRAIL must form a homotrimer to become biologically active. However, noncovalently associated TRAIL displays a high degree of sensitivity to degradation, which limits its therapeutic potential. To enforce trimerization of the recombinant protein, we developed a covalently linked TRAIL trimer (TR3) by genetic fusion. This molecular drug design conferred improved stability without altering the native killing ability of TRAIL. Target specificity was shown by blocking TR3 activity with soluble death receptor 5 (DR5-Fc). In addition, we have shown that TR3 is amenable to further, genetic modifications. The incorporation of additional functional domains to TR3, such as antibody fragments (scFvs) that allow for a more cell-specific delivery of the agent, is stoichiometrically controlled and inconsequential with regard to the bioactivity of TRAIL. As proof of this concept, TR3 activity was targeted to the mouse RBC membrane. TR3-decorated RBCs were effectively capable of target cell killing in a model of pancreatic cancer. TR3 represents a generally applicable platform tool to study basic mechanisms along the death receptor pathway. More importantly, the ability to target TR3 to a cell surface presents the opportunity to create a cancer-selective drug with fewer off-target toxicities and enhanced killing capacities.