Christoph Renner
Saarland University Medical School, Homburg, Germany

Generation of Biological Active Antibody Constructs for Cancer Therapy

Recent clinical trials in the area of non-Hodgkin’s lymphoma and breast cancer have changed the view on antibodies and initiated a revival of the concept of antibody-mediated tumor targeting. It is now accepted that antibody-based approaches will contribute to the field of tumor therapy. However, the clinical data reveal as well that the cytotoxic activity of monoclonal antibodies in general is too low to induce long-lasting remissions or even cure from the disease. Therefore, new treatment modalities on the basis of monoclonal antibodies have to be developed. Our group dedicated its work to approaches aimed at using antibodies to activate and redirect the defense mechanisms of the human immune system towards malignant cells. Almost 10 years ago, we started on bispecific antibodies (Bi-Mabs) for the treatment of Hodgkin’s lymphoma. Over the last years, two phase I/II clinical trials using an NK-cell activating bispecific antibody have been performed. Overall, 31 patients entered these studies and tolerated the treatment well without the occurrence of major adverse effects. An objective response rate of 30% (based on CT-scan criteria) with three complete remissions, six partial remissions and six patients with stable disease confirmed the efficacy of the treatment. However, in all patients responses were short lived with a range from 3-18 months. Because of the murine nature of the Bi-Mab, re-treatment of responding patients was limited by allergic reactions restricting the number of cycles administered to a maximum of four in some patients. With respect to the number of NK cells in peripheral blood before the start of the first treatment cycle, there was a trend toward higher NK cell counts in the responders as compared to the patients with stable disease or non-responders (p = 0.041). On average, NK cell counts were 2.04 times higher in responders than in patients with PD (95% CI, 1.08 –3.85). Inversely, the maximum NK cell activity in peripheral blood measured during the first treatment cycle was higher in the group of patients with PD as compared to patients with SD and responders (p = 0.006). The group of patients with SD appeared to be more similar to the group of responders than to the group of patients with PD with respect to NK cell counts and activity. The maximum proportion of NK cells with BiMAb bound during therapy was not significantly different between groups (p = 0.6). Encouraged by these results, we are in the process of generating humanized antibody fragments enabling us to continue clinical trials with less immunogenic antibodies.

To broaden our spectrum of biological active antibodies, we entered the field of immunocytokines and generated mammalian expression systems allowing high level production of recombinant antibodies and their fragments in CHO cells. As a first approach, we fused proapoptotic proteins such as tumor necrosis factor (TNF) or TNF-related-apoptosis-inducing-ligand (TRAIL) directly to the antibody using the antibody as a tumor-specific carrier for an endogenously occurring cytotoxic drug. Expression levels for the TNF construct ranged between 10-15 mg/l depending on the tumor specificity we aimed to target (non-Hodgkin’s lymphoma, Hodgkin’s lymphoma, renal cancer or tumor fibroblasts). Specific killing was observed down to the low pg/ml level with a LD50 activity between 1-10 ng/ml. Animal models are currently underway to confirm the activity in vivo and master cell banks are being established to proceed for GMP-production and subsequent clinical trials.