Multiple target-specific molecular imaging agents detect liver cancer in a preclinical model.

Liver cancer is the fifth most common cause of cancer deaths worldwide. Noninvasive diagnosis is difficult and the disease heterogeneity reduces the accuracy of pathological assays. Improvement in diagnostic imaging of specific molecular disease markers has provided hope for accurate and early noninvasive detection of liver cancer. However, all current imaging technologies, including ultrasonography, computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging, are not specific targets for detection of liver cancer. The aim of this study was to test the feasibility of injecting a cocktail of specific molecular imaging agents to noninvasively image liver cancer. The target-specific cocktail contained agents for imaging the neovasculature (RGD peptide), matrix metalloproteinase (MMP), and glucose transport ((18)F-fluorodeoxyglucose [(18)F-FDG]). Imaging studies were performed in liver cancer cells and xenograft models. The distribution of MMP at the intracellular level was imaged by confocal microscopy. RGD, MMP, and (18)F-FDG were imaged on tumor-bearing mice using PET, CT, X-ray, and multi-wavelength optical imaging modalities. Image data demonstrated that each agent bound to a specific disease target component. The same liver cancer xenograft contained multiple disease markers. Those disease markers were heterogenetically distributed in the same tumor nodule. The molecular imaging agents had different distributions in the whole body and inside the tumor nodule. All target-specific agents yielded high tumor-to-background ratios after injection. In conclusion, target-specific molecular imaging agents can be used to study liver cancer in vitro and in vivo. Noninvasive multimodal/multi-target-specific molecular imaging agents could provide tools to simultaneously study multiple liver cancer components.