TRAIL, a promising antitumor immuno-agent, exerted limited efficacy in clinical trials. The third disulfide loop of TGF-α (TGF3L peptide) with a very low affinity for EGFR has been reported to enhance the activity of fused antigens or cytokines. We wondered whether fusion of this peptide could enhance TRAIL activity and what the underlying mechanism for this enhancement would be. The TGF3L-TRAIL showed greatly enhanced cytotoxicity in a variety of cancer cell lines while spared normal cells unharmed. Typical apoptosis and cellular caspase activation were potently induced by TGF3L-TRAIL at the concentration levels corresponding to its cytotoxicity. TGF3L-TRAIL was able to activate both DR4 and DR5 the same as TRAIL did. It induced complete cell death in Colo205 through only one receptor when the other one was blocked, different from TRAIL-induced cell death (through DR4 dominantly). TGF3L-TRAIL cytotoxicity was not reduced in some cell lines even if both receptors are blocked simultaneously. Surprisingly, TGF3L-TRAIL self-assembled into stable polymers, which was responsible for its enhanced cytotoxicity. In human tumor xenograft mouse models, TGF3L-TRAIL showed anti-tumor activity similar to or better than TRAIL in different cancer cell types, consistent with its differing enhancement of cytotoxicity in vitro. Taken together, TGF3L fusion of TRAIL obviously enhances the anticancer activity of TRAIL by promoting assembly into polymers, which presents a novel fusion strategy for improving TRAIL function.