Background and objectives: We have set up a murine B lymphoma model stably expressing human CD20 and homing in lymph nodes in immunocompetent mice to study the mechanism of action of rituximab.
Design and methods: The B lymphoma line 38C13 was stably transduced with the human CD20 cDNA by retroviral infection and injected into syngeneic mice.
Results: The transduced 38C13-CD20(+) cells stably expressed human CD20 on 100% of cells. Rituximab alone did not inhibit 38C13-CD20+ cell growth but relocalized the human CD20 into lipid rafts and induced complement-mediated lysis in vitro. Inoculation of 4x10(3) 38C13-CD20(+) intravenously into syngeneic mice led to the development of tumor masses in the spleen, bone marrow and lymph nodes, detectable from day 15 by polymerase chain reaction (PCR) analysis, and with a median survival of 21-24 days. Treatment with 250 mg rituximab i.p. given 1-10 days after tumor inoculation cured 100% of animals, with disappearance of tumor documented by immunohistochemistry and PCR analysis. Depletion of both NK cells and neutrophils did not affect the therapeutic activity of rituximab in vivo. Similarly, removal of phagocytic macrophages using clodronate-liposomes did not modify the capacity of rituximab to control tumor growth. In contrast, the protective activity of the antibody was completely abolished after complement depletion with cobra venom factor. Complement was also required when cells were inoculated subcutaneously in nude mice.
Interpretation and conclusions: These data demonstrate that complement is required for the therapeutic activity of rituximab in vivo in a murine model of B-cell lymphoma, independently of its localization in lymph nodes or subcutaneously.