Specific binding of molecularly targeted agents to pancreas tumors and impact on observed optical contrast

In optical imaging it is thought that optimum tumor contrast can be achieved with the use of small-labeled molecular tracers that have high affinity to their targets and fast clearance rates from the blood stream and healthy tissues. An example of this is fluorescently tagged EGF to monitor the molecular activity of tumors, such as pancreatic cancer. Extensive fluorescence contrast analysis for fluorescence molecular tomography has been performed on the AsPC-1 pancreas tumor, grown orthotopically in mice; yet, the binding dynamics of the EGF-fluorescent agent in vivo is not completely known. The bulk pancreatic tumor displays 3:1 contrast relative to the normal pancreas at long times after injection; however, even higher levels of fluorescence in the liver, kidney and intestine suggest that molecular specificity for the tumor may be low. Mice were administered a fluorescently labeled EGF agent and were sacrificed at various time points post-injection. To analyze the amount of specific binding at each time point frozen tissue samples were fluorescently imaged, washed with saline to remove the interstitially distributed contrast agent, and then imaged again. This technique demonstrated that approximately ~10% of the molecular target was firmly bound to the cell, while 90% was mobile or unbound. This low binding ratio suggests that the contrast observed is from inherent properties of the tumor (i.e. enhanced permeability and retention effect) and not from specific bound contrast as previously anticipated. The use of EGF contrast agents in MRI-guided fluorescence tomography and the impact of low binding specificity are discussed.