Abstract
Background: CD22 has emerged as an attractive target for the treatment of B-cell malignancies because (i) CD22 shows high surface expression on most mature and precursor B-cell malignancies, (ii) it is rapidly internalized after ligand binding, and (iii) it is readily replenished after internalization due to a substantial intracellular pool. CD22 has been engaged therapeutically using CAR-T cells (Frey, 2018), immunotoxins (Wayne, 2018), or Antibody-drug-conjugates (Kantarjian, 2016). Preclinical data, however, has exclusively been generated using immune compromised mice which is a major drawback of current animal models.
Goal: We aimed to establish a syngeneic, murine lymphoma model expressing the human CD22 antigen under physiologic promoter control to test CD22-targeted therapies in an immune competent background.
Methods/Results: A CD22 chimera (h/mCD22) was designed as human on the outside and murine on the inside. Engaged by the CD22-targeted immunotoxin Moxetumomab pasudotox (Moxe) in vitro, h/mCD22 successfully transports Moxe which exclusively targets the human CD22 to the cytosol of murine cells resulting in a dose dependent cytotoxicity. By cross-breeding BL6 mice expressing h/mCD22 as a transgene (BL6tm(h/mCD22)Eng) and the BL6λ-myc mouse strain, we generated mice that spontaneously develop h/mCD22-positive lymphoma. Three primary lymphomas were isolated from distinct mice and each serially transplanted into BL6tm(h/mCD22)Eng mice. Stable engraftment and tumor cell growth was established after subcutaneous (sc) as well as intravenous (iv) injection. Tumor cells isolated from sc tumors, however, were substantially smaller than tumor cells isolated from bone marrow (BM), Spleen (SPL), or lymph nodes (LN) after iv injection. Correlating with the approximately 2-fold redcued cell surface of the smaller sc tumor cells, surface CD22 was reduced by 2-fold. Importantly, sc tumors were infiltrated by less than 1% immune cells, while myc-driven lymphoma in men commonly present with substantial tumor-infiltrating immune cells. Resembling human disease, the systemically growing tumors after iv injection were infiltrated by 20% myeloid cells in BM, by 5% in LN, and by 10 % in SPL and were infiltrated by 0.5% T-cells in BM, by 30% in LN, and by 7% in SPL. Testing Moxe against the three distinct primary murine B-NHL models in vivo, we found Moxe to be more active, the longer high blood levels were maintained, reflecting results from studies in immune compromised models (Müller, 2016). Moxe given as 4 doses intraperitoneally every day for 5 days achieved stable disease in one and reduced tumor burden by more than 10-fold and by more than 50-fold in two other systemic mouse models.
Conclusion: By cross-breeding h/mCD22-transgenic and λ-myc expressing mouse strains, we generated primary aggressive B-cell lmyphoma expressing h/mCD22. The lymphomas engraft in h/mCD22-transgenic BL6 mice and iv but not sc injected tumors recapitulate the immune infiltration found in human myc-translocated B-NHL. Our unique models provide a valuable platform to test treatment modalities targeting human CD22 in more relevant models of disease.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.