Ex vivo and in vivo evaluation of overexpressed VLA-4 in multiple myeloma using LLP2A imaging agents

Very-late-antigen-4 (VLA-4, α₄β₁ integrin, CD49d/CD29) is a transmembrane adhesion receptor that plays an important role in cancer and immune responses. Enhanced VLA-4 expression has been observed in multiple myeloma (MM) cells and surrounding stroma. VLA-4 conformational activation has been associated with MM pathogenesis. VLA-4 is a promising MM imaging and therapeutic biomarker. Methods: Specificity of ⁶⁴Cu-LLP2A (⁶⁴Cu-CB-TE1A1PPEG₄- LLP2A), a high-affinity VLA-4 peptidomimetic-based radiopharmaceutical, was evaluated in α₄ knock-out mice and by competitive blocking in wild-type tumor-bearing mice. ⁶⁴Cu-LLP2A PET/CT (static and dynamic) imaging was conducted in C57BL6/KaLwRij mice bearing murine 5TGM1-GFP syngeneic tumors generated after intravenous injection via the tail. Blood samples were collected for serum protein electrophoresis. Bone marrow and splenic cells extracted from tumor-bearing and control mice (n = 3/group) were coincubated with the optical analog LLP2A-Cy5 and mouse B220, CD4, Gr1, and Mac1 antibodies and analyzed by fluorescence-activated cell sorting. Human radiation dose estimates for ⁶⁴Cu-LLP2A were extrapolated from mouse biodistribution data (6 time points, 0.78 MBq/animal, n = 4/group). Ten formalin-fixed paraffin-embedded bone marrow samples from deceased MM patients were stained with LLP2A-Cy5. Results: ⁶⁴Cu-LLP2A and LLP2A-Cy5 demonstrated high specificity for VLA-4-positive mouse 5TGM1-GFP myeloma and nonmalignant inflammatory host cells such as T cells and myeloid/monocytic cells. Ex vivo flow cytometric analysis supported a direct effect of myeloma on increased VLA-4 expression in host hematopoietic microenvironmental elements. SUVs and the number of medullar lesions detected by ⁶⁴Cu-LLP2A PET corresponded with increased monoclonal (M) protein (g/dL) in tumor-bearing mice over time (3.29 ± 0.58 at week 0 and 9.97 ± 1.52 at week 3). Dynamic PET with ⁶⁴Cu-LLP2A and ¹⁸F-FDG demonstrated comparable SUV in the prominent lesions in the femur. Human radiation dose estimates indicated urinary bladder wall as the dose-limiting organ (0.200 mGy/MBq), whereas the dose to the red marrow was 0.006 mGy/MBq. The effective dose was estimated to be 0.017 mSv/MBq. Seven of the ten human samples displayed a high proportion of cells intensely labeled with LLP2A-Cy5 probe. Conclusion: ⁶⁴Cu-LLP2A and LLP2A-Cy5 demonstrated binding specificity for VLA-4 in an immune-competent murine MM model. ⁶⁴Cu-LLP2A displayed favorable dosimetry for human studies and is a potential imaging candidate for overexpressed VLA-4.