Oral Presentation Advances in Neuroblastoma Research Congress 2016

A CD56 (NCAM1) targeting antibody-drug conjugate is potently effective in preclinical models of high-risk neuroblastoma (#25)

Robyn T Sussman 1 , Michael Randall 1 , Kevin Huang 1 , Kristopher Bosse 1 2 , Yang Feng , Yanping Wang , Zhongyu Zhu 3 , Dimiter S Dimitrov 3 , John M Maris 1 2
  1. Children's Hospital of Philadelphia, Philadelphia, PENNSYLVANIA, United States
  2. Pediatrics, University of Pennsylvania, Philadelphia, PA, USA
  3. Cancer and Inflammation Program, Center for Cancer Research, National Cancer Institute, Frederick, MD, USA

Background: The GD2-targeting monoclonal antibody dinutuximab was the first FDA approved immunotherapy specifically for high-risk neuroblastoma. While dinutuximab therapy improved event-free survival rates, therapy is associated with severe pain caused by on-target activity on nerve cells, highlighting the need to consider other cell surface molecules as candidate targets for immunotherapy. CD56 is robustly over-expressed in many cancers including neuroblastoma.

 

Methods: We developed a fully human IgG1 targeting CD56 (m906-IgG) following phage-display-based isolation and optimization of CD56-specific binders. Using antibody glycan sites on m906-IgG1 for conjugation, we synthesized an antibody drug conjugate (ADC) with the DNA intercalating agent pyrrolobenzodiazepine dimer (m906-PBD) and tested for anti-neuroblastoma activity in both in vitro and in vivo preclinical models.

 

Results: Using immunofluorescence and flow cytometry, we showed that m906-IgG is rapidly internalized in a panel of 20 neuroblastoma cell lines. We next demonstrated potent cytotoxicity across this cell line panel (median IC50 = 1.48nM, range 10.5pM to 4.1nM) Somatic TP53 mutation was a biomarker for resistance to the ADC in the panel (p=0.03; median IC50 .for cell line models with wild-type TP53 = 211pM with a range from 10.5pM to 1.5nM vs median IC50 for mutated lines = 260nM, range 2.7nM to 4.1uM). Antibody-competition studies confirmed the specificity of m906-PBD. In an initial dose finding experiment using the NB-1643 patient-derived xenograft model and a twice weekly for a total of 4 doses schedule, we showed significant tumor regression at 1 mg/kg/dose, with eventual tumor regrowth at a median of 40 days. At 3mg/kg/dose, there were sustained regressions without regrowth out to 110 days.

 

Conclusions: The CD56-targeting m906-PBD ADC shows potent cytotoxicity in models of high-risk neuroblastoma.