Breakthroughs have been achieved in cancer immunotherapies recently. Some patients experienced a durable remission or even cure by the treatments activating patients’ own T cells. However, these single target technologies have failed to result in any therapeutic effect in a large percent of patients, and evidence suggests that further advances depend on the effective strategy for coping with cancer heterogeneity and dynamics. We hypothesize that minicircles, a class of optimized DNA vectors, are able to cope with these problems; that is to apply minicircles to express a panel of engineered antibodies capable killing all cell lineages in a cancer by inducing antibody-dependent cellular cytotoxicity or retargeting CTLs to become anticancer T cells. To prove this concept, we constructed a minicircle to express a bispecific antibody (BsAB) capable of binding CD3 on T cell and CD20 on B-lymphoma cell simultaneously, and tested if it could function like its BsAb counterpart blinotumomab in treating B-lymphoma. Six to eight week-old female SCID/Beige mice were inoculated with 2×5e5 Raji cells through tail vein injection, and 4 μg of the minicircle was injected into the mouse liver via hydrodynamic technique 24 hours later. Selected mice received 3 intravenous doses of human DC-CIK cells, 5e6 each, every other day starting 2 days after Raji cell inoculation. All the 14 Raji cell-bearing mice receiving no further treatment died in 3 weeks, the 10 mice that received DC-CIDK as well died in 7 weeks, while the10 mice received both MC. CD3xCD20 and DC-CIK survived for additional 5 weeks. Histology examination demonstrated human CD3-positive T cells and CD20-positive B-lymphoma cells in spleens. In another fine-tuning experiment, we imaged the mice receiving the same number of Raji cell expressing luciferase gene when they demonstrated paralyzed rear legs, the typical syndrome indicating the late stage of this disease. We observed strong luciferase signal along the hips and spinal of the mice. Taken together, these data demonstrated that we have successfully established the mouse model of human B-lymphoma, and that minicrcle was able to generate therapeutic level of BsAb in vivo. Further experiments are undergoing to characterize the minicircle-BsAb anti-cancer system. As compared to the virus vectors, minicircle is safer, more convenient to make, easier to target multiple cancer cell lineages simultaneously, our efficacy data support the notion that minicircles have the potential to serve as powerful immunotherapy vectors in fighting human cancer.