Abstract 5444: Development of a noninvasive assay to determine drug concentration in tumor during hsp90 inhibitor therapy

• Published in: Cancer research (Chicago, Ill.) Ill.), 2015-08, Vol.75 (15_Supplement), p.5444-5444
• Main Authors: Taldone, Tony, Pillarsetty, Nagavarakishore, Dunphy, Mark PS, Gerecitano, John F., Caldas-Lopes, Eloisi, Beattie, Brad, Peter, Radu I., Kang, Yanlong, Rodina, Anna, Yan, Pengrong, DaGama Gomes, Erica M., Bolaender, Alexander, Pressl, Christina, Punzalan, Blesida, Ku, Anson, Ku, Thomas, Shah, Smit, Uddin, Mohammad, Chen, Mei H., Santos, Elmer, Koziorowski, Jacek, Corben, Adriana, Modi, Shanu, Jhaveri, Komal, Lin, Oscar, Vakiani, Efsevia, Janjigian, Yelena, Zanzonico, Pat, Hudis, Clifford, Larson, Steven M., Lewis, Jason S., Chiosis, Gabriela
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2015-5444

As molecularly targeted agents assume a more prominent role in anticancer therapy there is a growing need to determine in a noninvasive manner whether the target is being engaged and to what extent such drug-target binding results in desirable effects. We address this need in the context of Hsp90, a target of significant value and one in critical need for such assessment tools, by combining a novel chemical tool selective for tumor Hsp90 with PET imaging and mathematical modeling. The chemical tool is [124I]-PU-H71, the iodine-124 radiolabeled analog of the potent Hsp90 inhibitor PU-H71, which can be administered in tracer quantities for PET imaging. The resulting diagnostic, PU-PET, has been optimized and validated preclinically in mouse models of cancer and then translated to the clinic. The exquisite design of this assay is based on three essential concepts as it relates to the target (Hsp90) as well as to the PET tracer (124I-PU-H71). First, the target is “oncogenic” Hsp90 and has been shown by numerous biochemical and pharmacokinetic studies to have a strong affinity for inhibitors and a very low koff resulting in selective and prolonged retention in tumor. Secondly, the tracer incorporates a 124I in place of the naturally occurring 127I in the structure of PU-H71 and therefore there is no change in the chemical structure. This feature in a PET tracer intended as a companion diagnostic is unprecedented and ensures that the PK properties are identical to the therapeutic agent (PU-H71). Finally, the radionuclide 124I has a four-day half-life and thus is well-suited to monitor the extended tumor retention profile observed for Hsp90 inhibitors. We here demonstrate that this PET assay informs on Hsp90 targeting in individual tumors in real time and provides accurate tumor drug concentrations for at least four chemically distinct Hsp90 drugs. In contrast, we find that plasma pharmacokinetics is not predictive of intratumor parameters and therefore provides limited value in estimating target engagement. Using PU-PET we demonstrate that at least one Hsp90 inhibitor exhibits tumor targeting and retention in humans, delivering and retaining therapeutic, micromolar, concentrations at safe doses. PU-PET is currently being evaluated in Phase 0/1 (NCT01269593) clinical trials as a noninvasive companion diagnostic to determine intratumoral concentration as well as to identify those patients who would best benefit from Hsp90 inhibitor therapy. This diagnostic assay i