Abstract
1208
Objectives: There is an unmet clinical need for less invasive methods to detect and quantify colitis. We previously reported the detection of gut inflammation in colitic mice using 64Cu-labeled FIB504.64, an antibody to the β7 integrin monomer which is expressed by lymphocytes involved in colitis. The recruitment and retention of lymphocytes to regions of colitis is via the interaction of lymphocyte-expressed integrins incorporating β7 with mucosal addressin cell adhesion molecule (MAdCAM) expressed on the surface of mucosal cells. To further investigate this physiological process and its potential for use in colitis imaging, we constructed an antibody that incorporates MAdCAM, specifically its D1D2 domains, and used it to image colitis.
Methods: The MAdCAM immunoprotein (R-MAdCAM) and its non-binding control, Mutant-MAdCAM (M-MAdCAM), were labeled with 64Cu using NOTA. The biodistribution of R- and M-MAdCAM was measured in healthy and piroxicam-induced colitic C57/bl-IL10KO mice (n = 4-5). Small animal PET data were obtained at 1, 4, and 24 h post-injection (p.i.), and an ex vivobiodistribution study was carried out after the final imaging session. The degree of colitis was assessed by measuring colonic density.
Results: PET image data showed higher uptake in liver than other tissues, and some uptake throughout the gut. In contrast to our previous immunoPET studies of colitis, there were no foci of tracer uptake in the gut in colitic mice. Ex vivobiodistribution data confirmed gut uptake. For R-MAdCAM, large colon uptake was 5.65±0.72% ID/g in healthy mice and 5.63±0.94% ID/g in colitic mice (n = 4/5, mean±standard deviation, p>0.05). For M-MAdCAM, large colon uptake in control mice was 5.29±1.11% ID/g and 5.05±1.0% ID/g in colitic mice. Colon densities for control and piroxicam-treated groups were 2.15±0.22 and 4.24±0.80 mg/mm (means of n=10±SD, p < 0.001), confirming the presence of disease. The data were examined in greater detail, and tracer uptake for 3 tissues (large intestine, small intestine, and stomach) was plotted against colonic density. These plots indicated that for the animals with colitis, uptake of target-binding R-MAdCAM protein increased with colonic density, while for the control animals tracer uptake decreased with colonic density. The slopes of the plots for control mice were negative (-7.39 (SE 4.94), -6.60 (6.12), and -8.32 (5.07)) for the large intestine, small intestine, and stomach, respectively, and positive for colitic mice (0.88 (0.57), 1.61 (0.71), and 0.36 (0.59) for the same tissues). Interestingly, the differences between the slopes of the colitis and control plots (colitis-control) were similar across these three tissues: 8.27 (SE 4.98), 8.20 (6.16), and 8.68 (5.11). The similarity of the colitis-control differences between the slopes was confirmed by fitting a model using the data for all tissue types. The common estimate based on this model was 8.38 (SE 3.41), and the p-value was 0.014 indicating >95% confidence that there is a difference between the slopes of the two populations.
Conclusions: These results suggest that uptake of 64Cu-labeled R-MAdCAM by the gut correlates with colonic density and thus has the potential to be used as a PET tracer for the evaluation of colitis.