Gamma imaging of atherosclerotic lesions: the role of antibody affinity in in vivo target localization.

Background: Monoclonal antibodies are attractive agents for noninvasive localization of various cardiovascular disorders. Because proliferating intimal smooth muscle cells are important components of atherosclerotic lesions, radiolabeled antibody Z2D3 specific for proliferating smooth muscle cells has been used for immunoscintigraphic localization of experimental atherosclerotic lesions. This study was undertaken to assess the role of antibody affinity in optimization of immunoscintigraphic localization of these lesions. Z2D3 belongs to the immunoglobulin (Ig) M class of antibodies. For immunoscintigraphic studies, attempts were made to prepare F(ab')2 or Fab fragments from the parent cell line. Fragmentation of Z2D3-IgM or its two subclones (B7 and 2B12) was not possible; therefore the parent hybridoma line was subjected to class switching. Cell lines 5C5 and 3E5 secreted antibody of the IgG1 subclass. The Z2D3-IgG1 antibodies were enzymatically digested to provide F(ab')2. Because of a tenfold loss of immunoreactivity of these class-switched antibodies, the parent clone was subsequently genetically engineered to obtain a mouse/human chimeric antibody with human IgG1 constant region. F(ab')2 of Z2D3-73.30 chimeric antibody retained the immunoreactivity relative to the original Z2D3-IgM. Radiolabeled murine and chimeric F(ab')2 fragments were used to assess the role of affinity in gamma scintigraphic visualization of experimental atherosclerotic lesions.

Methods And Results: Experimental atherosclerotic lesions were induced in 19 rabbits by abdominal aortic balloon deendothelialization followed by a hyperlipidemic diet for 12 weeks. 111In-labeled chimeric high-affinity Z2D3 F(ab')2 fragments (111In-Hi.aff Z2D3) were administered in four animals. Uptake was compared with 111In-labeled F(ab')2 of nonspecific human IgG1 (n = 4), murine low-affinity Z2D3-5C5 (111In-Lo.aff Z2D3; n = 4), nonspecific murine IgG1 monoclonal antibodies (n = 4), and 123I-labeled murine low-affinity Z2D3-3E5 (n = 3). Atherosclerotic lesions were visualized 48 hours after administration of the chimeric Hi.aff-Z2D3 antibody in all animals. Lesions were not visualized in rabbits injected with Lo.aff-Z2D3 or murine or human nonspecific antibodies. Mean percent injected dose per gram in the lesion was significantly higher with the 111In-Hi.aff-Z2D3 (0.112% +/- 0.024%) compared with 111In-Lo.aff-Z2D3 (0.037% +/- 0.005%; p = 0.03), human nonspecific (0.027% +/- 0.004%; p = 0.01), or murine nonspecific antibodies (0.006% +/- 0.001%; p = 0.0004). Nonspecific activity in unballooned thoracic aortic segments (normal) was comparable in the 111In-Hi.aff-Z2D3 (0.019 +/- 0.003) and the 111In-Lo.aff-Z2D3 (0.011% +/- 0.005%; p = 0.3) antibodies. The lesion activities of the Lo.aff-Z2D3 labeled with 111In (0.037 +/- 0.005) or 123I (0.034 +/- 0.007; p = 0.71) were similar regardless of the radioisotopes used for labeling.

Conclusions: Our study demonstrates that the specificity of an antibody for the target antigen in the atheroma is a necessary condition for in vivo targeting. However, high enough affinity of an antibody is an essential component for noninvasive diagnostic visualization of experimental atherosclerotic lesions.