Many nanoparticle based imaging agents (including those developed by the Co-PI) are in preclinical use [1], and are currently being prepared for clinical use. Of particular interest are molecularly targeted nanoparticle imaging agents that combine the benefits of nanoencapsulation [2] to the molecular specificity of ligand-targeted imaging [3]. The nanoparticles are designed to have extended blood pool residence time, facilitating multiple opportunities to contact the target site, and thus binding to the target. A key obstacle to this approach is the background signal due to the presence of unbound imaging agent. A means of clearing unbound agent from the circulation "on-command", leaving only the bound particles is needed. We have constructed a lipid-PEG linkage using a dithiobenzyl (DTB) linking group that is cleavable by weak thiols such as cysteine and glutathione [4]. In the current project, we will therefore create Stealth nanoparticles with a DTB-attached polyethylene glycol coating. The targeting ligand will be attached by a non-cleavable PEG. Thus, upon injection, the targeted particles will have ample opportunity to bind to target, and once sufficient binding has been realized, a follow-up cysteine injection will cause the rapid clearance of all unbound agent from the blood pool. The remaining contrast signal will thus be uniquely due to the bound agent, enabling improved visualization. The cleavable strategy may be applicable to a variety of nanoparticles. We will apply this technique to image a clinically relevant target, namely the somatostatin receptor (SSTR). SSTR imaging today uses radiolabeled peptide analogues of somatostatin to clinically evaluate neuroendocrine tumors such as carcinoid and lung tumors. Significant shortcomings of this approach are its poor spatial resolution and its inability to visualize normal anatomical structures and landmarks. A SSTR imaging technique that maintains high spatial resolution, sensitivity, and penetration depth, while also visualizing normal anatomical structures would therefore be of enormous interest. A further attraction of SSTR-targeted particles is the potential in the future to codeliver therapeutic agents along with the imaging agents, thus facilitating simultaneous imaging and therapy with the same high degree of specificity. Metastatic carcinoid and lung tumors are currently very difficult to cure and would be immediately accessible by these techniques. In the near future, one also envisions the ability to image and eventually treat a variety of tumors that are

made to express somatostatin receptors, for example, by gene transfer of the receptor.

1. Construct Nanoparticles:

T o construct "on-command" Stealth nanoparticles, targeted by Octreotide to the Somatostatin receptor.

2. Test Localization:

T o test the localization of these nanoparticles in tumors overexpressing the Somatostatin receptor.

3. Test on-command cleavability:

T o test the on-command clearance of these nanoparticles from the blood pool, and demonstrate the enhancements in contrast at the target achievable by the on-command clearance technique.

Associated animal protocol at UT-MDACC: Molecular Imaging (mouse) #080107632