Radiohybridization PET imaging of KRAS G12D mRNA expression in human pancreas cancer xenografts with [(64)Cu]DO3A-peptide nucleic acid-peptide nanoparticles.

There is a compelling need to image pancreas cancer at an early stage. Human pancreas cancer cells display elevated levels of KRAS protein due to high copy numbers of KRAS mRNA, and elevated levels of insulin-like growth factor 1 receptor (IGF1R) due to overexpression of IGF1R mRNA. Therefore we hypothesized that pancreas cancer could be detected in vivo with a single probe that targets both KRAS mRNA and IGF1R. Because positron emission tomography (PET) is a sensitive imaging technique, we designed a probe incorporating the positron-emitting nuclide (64)Cu. The KRAS-specific hybridization probe consisted of 1,4,7-tris(carboxymethylaza)cyclododecane-10-aza-acetyl (DO3A) on the N-terminus of a peptide nucleic acid (PNA) hybridization sequence (GCCATCAGCTCC) linked to a cyclized IGF1 peptide analog (d-Cys-Ser-Lys-Cys) on the C-terminus, for IGF1R-mediated endocytosis. A series of such KRAS radiohybridization probes with 0, 1, 2 or 3 mismatches to KRAS G12D mRNA, including exact matches to wild type KRAS mRNA and KRAS G12V mRNA, along with a double d(Ala) replacement IGF1 peptide control, were assembled by continuous solid phase synthesis. To test the hypothesis that KRAS-IGF1 dual probes could specifically image KRAS mRNA expression noninvasively in human IGF1R-overexpressing AsPC1 pancreas cancer xenografts in immunocompromised mice, [(64)Cu]PNA radiohybridization probes and controls were administered by tail vein. The [(64)Cu]KRAS-IGF1 radiohybridization probe yielded strong tumor contrast in PET images, 8.6 +/- 1.4-fold more intense in the center of human pancreas cancer xenografts than in the contralateral muscle at 4 h post-injection. Control experiments with single base KRASmismatches, an IGF1 peptide mismatch, and a breast cancer xenograft lacking KRAS activation yielded weak tumor contrast images. These experiments are consistent with our hypothesis for noninvasive PET imaging of KRAS oncogene expression in pancreas cancer xenografts. Imaging oncogene mRNAs with radiolabel-PNA-peptide nanoparticles might provide specific genetic characterization of preinvasive and invasive pancreas cancers for staging and choice of therapy.

Receptor-mediated internalization of chelator-PNA-peptide hybridization probes for radioimaging or magnetic resonance imaging of oncogene mRNAs in tumours.

Early external detection of cancer gene activity might enable early treatment of cancer and might reduce cancer mortality. We hypothesized that oncogene mRNA overexpressed at thousands of copies per malignant cell in a zone of transformed cells could be imaged externally by scintigraphic imaging, PET (positron emission tomography) or MRI (magnetic resonance imaging) with PNA (peptide nucleic acid) hybridization probes that include chelators for metal cations and a cyclized peptide analogue of IGF-1 (insulin-like growth factor 1), D(Cys-Ser-Lys-Cys), to mediate internalization by IGF1R (IGF-1 receptor) overexpressed on cancer cells. We observed that human MCF7 breast cancer cells that overexpress IGF1R efficiently internalized fluorescein-chelator-PNA-D(Cys-Ser-Lys-Cys) to the cytoplasm, but not with D(Cys-Ala-Ala-Cys). Scintigraphic imaging of MCF7 xenografts in immunocompromised mice revealed that CCND1 and MYC [(99m)Tc]chelator-PNA-D(Cys-Ser-Lys-Cys) probes yielded xenograft. PET imaging with [(64)Cu]chelator-PNA-D(Cys-Ser-Lys-Cys) yielded stronger signals. Scintigraphic imaging of human AsPC1 pancreas cancer xenografts with [(99m)Tc]chelator-KRAS PNA-D(Cys-Ser-Lys-Cys) yielded strong xenograft signals. Stronger xenograft image intensities were obtained by PET imaging of [(64)Cu]chelator-KRAS PNA-D(Cys-Ser-Lys-Cys). MRI required extension of chelator-polydiamidopropanoate dendrimers from the N-termini of the PNA probes to increase the number of contrast paramagnetic gadolinium (III) cations per probe. These results provide a basis for detection of oncogene activity in tissues from outside the body by hybridization with metal-chelator-PNA-peptides that are selectively internalized by cancer cells.

Noninvasive imaging of tumor hypoxia in rats using the 2-nitroimidazole 18F-EF5.

Tumor hypoxia is an important prognostic indicator for cancer therapy outcome. EF5 [2-(2-nitro-1[ H]-imidazol-1-yl)- N-(2,2,3,3,3-pentafluoropropyl)-acetamide] has been employed to measure tumor hypoxia in animals and humans using immunohistochemical methods. EF5 is a lipophilic molecule designed to have a very uniform biodistribution, a feature of obvious benefit for use in PET imaging. The present study represents the first demonstration of noninvasive PET imaging of rat tumors using fluorine-18 labeled EF5. Because of the small tumor size, partial volume effects may result in underestimation of concentration of the compound. Therefore, validation of the PET data was performed by gamma counting of the imaged tissue. The tumor models studied were the Morris 7777 (Q7) hepatoma (n=5) and the 9L glioma (n=2) grown subcutaneously in rats. Our previous studies have demonstrated that early passage 9L tumors are not severely hypoxic and that Q7 tumors are characterized by heterogeneous regions of tumor hypoxia (i.e., Q7 tumors are usually more hypoxic than early passage 9L tumors). The seven rats were imaged in the HEAD Penn-PET scanner at various time points after administration of 50-100 micro Ci (18)F-EF5 in 30 mg/kg carrier nonradioactive EF5. The carrier was used to ensure drug biodistribution comparable to prior studies using immunohistochemical methods. (18)F-EF5 was excreted primarily via the urinary system. Images obtained 10 min following drug administration demonstrated that the EF5 distributed evenly to all organ systems, including brain. Later images showed increased uptake in most Q7 tumors compared with muscle. Liver uptake remained relatively constant over the same time periods. Tumor to muscle ratios ranged from 0.82 to 1.73 (based on PET images at 120 min post injection) and 1.47 to 2.95 (based on gamma counts at approximately 180 min post injection). Tumors were easily visible by 60 min post injection when the final tumor to muscle ratios (based on gamma counts) were greater than 2. Neither of the 9L tumors nor the smallest Q7 tumor met this criterion, and these tumors were not seen on the PET images. These preliminary results suggest that (18)F-EF5 is a promising agent for noninvasive assessment of tumor hypoxia. Plans are underway to initiate a research project to determine the safety and preliminary evidence for the efficacy of this preparation in patients with brain tumors.