PET imaging of oncogene overexpression using 64Cu-vasoactive intestinal peptide (VIP) analog: comparison with 99mTc-VIP analog.

The purpose of this study was to assess the feasibility of PET imaging of oncogene VPAC1 receptors overexpressed in human breast cancer cells. METHODS: Vasoactive intestinal peptide (VIP) analog (TP3982) was synthesized to harbor a carboxy-terminus lysine (Lys) residue separated from VIP-asparagine (Asn(28)) by 4-aminobutyric acid (Aba) as a spacer. Lys was derivatized with diaminopropionic acid coupled to a pair of dibenzoylthioglycolic acid residues as protecting groups. The analog was labeled with (64)Cu at pH 9 ((64)Cu-TP3982) and (99m)Tc at pH 12 ((99m)Tc-TP3982). (99m)Tc-TP3982 and VIP derivatized with Aba-GAGG and labeled with (99m)Tc ((99m)Tc-TP3654) were used as reference agents. Smooth muscle relaxivity assays performed with each derivative and compared with unaltered VIP(28) demonstrated functional integrity. In vitro stability of (64)Cu-TP3982 was determined by challenging the complex with 100-mol excess of diethylenetriaminepentaacetic acid (DTPA), human serum albumin (HSA), and cysteine. In vivo stability was determined in urine and serum for up to 24 h. The mass of the Cu-TP3982 complex was determined by mass spectrometry. Human T47D breast tumor xenografts were grown in athymic nude mice. Planar scintigraphic imaging was performed at 4 and 24 h after the intravenous administration of (99m)Tc-TP3982 and (99m)Tc-TP3654 and PET imaging was performed using a small animal MOSAIC PET scanner, also at 4 and 24 h after injection of (64)Cu-TP3982. Tissue-distribution studies were also performed. In a separate experiment, receptors were blocked by intravenous injection of authentic VIP(28) 30 min before the administration of (64)Cu-TP3982 and tissue distribution was examined. RESULTS: (64)Cu-TP3982 labeling yields were 98% +/- 1.2% and those for (99m)Tc-TP3982 and (99m)Tc-TP3654 were 98.2% +/- 1.1% and 97% +/- 1.6%, respectively. The biologic activity of both VIP analogs was uncompromised. When (64)Cu-TP3982 was challenged with 100-mol excess of DTPA, HSA, or cysteine, >98% radioactivity remained as (64)Cu-TP3982. In vivo, >98% of (64)Cu circulating in plasma remained as (64)Cu-TP3982. Of the (64)Cu excreted in urine 4, 20, and 24 h after injection, >98%, 89.9% +/- 0.9%, and 85% +/- 3%, respectively, were bound to TP3982. The mass of Cu-TP3982 as determined by surface-enhanced laser desorption/ionization time of flight (SELDI-TOF) was 4,049.7 Da. Four hours after receptor blocking with VIP(28), there was a significant reduction in uptake of all tissues except in the liver. With (64)Cu-TP3982, the 4-h postinjection tumor uptake was 10.8 +/- 2.1 %ID/g versus 0.5 +/- 0.02 %ID/g and 0.24 +/- 0.08 %ID/g for (99m)Tc-TP3982 and (99m)Tc-TP3654, respectively. Twenty-four hours after injection, the corresponding numbers were 17 +/- 0.7 %ID/g, 0.77 +/- 0.1 %ID/g, and 0.23 +/- 0.1 %ID/g. The severalfold greater uptake (21.2-74) of (64)Cu-TP3982 is attributable to the in vivo stability of the agent. CONCLUSION: The results suggest that the uncompromised biologic activity and the significantly greater tumor uptake of (64)Cu-TP3982, combined with the high sensitivity and enhanced resolution of PET imaging, make (64)Cu-TP3982 highly desirable for further studies in PET imaging of oncogene receptors overexpressed in breast and other types of cancers.