healthmedicinet

Due to its high overexpression, GRP-R is an attractive target receptor for imaging and therapy of several cancers. Different peptides based on the wild-type BN [7–14] sequence, with agonist or antagonist properties, have been described in recent years and have been characterized as imaging tracers after their labeling with gamma or positron emitters. However, several different approaches for peptide labeling and other modifications, such as introduction of spacers between the targeting moiety and the chelator group, have been described. This does not allow to accurately compare and evaluate the importance of the peptide structure for binding and imaging properties. We have attempted to address this issue by comparing eight BN[7–14] peptide conjugates having the same spacer and metal chelator at N-terminus.

Based on data available in the literature, we have selected and characterized eight DOTA coupled peptide sequences that are bombesin analogs (Fig. 1). These publications served as basis for peptide design [15, 19, 26, 28, 29]. For the BN2 peptide, Leu¹³-Met¹⁴ has been substituted with Cha¹³-Nle¹⁴, a previously described strategy aimed at improving peptide stability in vivo [13]. The BN3 analog derives from the BN2 analog and contains an additional glutamic acid residue on the C-terminus that was introduced in an attempt to improve solubility. In BN4, the dipeptide Leu¹³-Met¹⁴ has been substituted with Sta¹³-Leu¹⁴ in an effort to improve in vivo stability to aminopeptidase activity. The Leu¹³-to-Sta¹³ substitution has also been shown to provide antagonistic properties to the peptide [17]. Peptide BN5 is a further modification of BN4 in which the Gly¹¹ residue has been substituted with -methylglycine to improve stability of the Gly–His bond to carnosinase enzymatic cleavage in vivo. The BN6, BN7, and BN8 peptides are modifications of the BN4, BN2, and BN5 peptides, respectively, in which a DPhe has been introduced at the N-terminus in order to increase binding affinity as previously described [11]. A hydrophilic spacer, formed by four ethoxylic groups (PEG4) was interposed between the peptide and the DOTA chelator. This approach has been previously described with DOTA coupled bombesin derivatives, and this particular spacer length was chosen based on evidence showing improved pharmacokinetic properties [22]. Peptides were synthesized in solid phase with Fmoc/tBu chemistry and obtained in high purity after RP-HPLC chromatography.

The radiolabeled peptide derivatives show no significant differences in affinity and binding capacity on PC-3 cells, with the exception of the BN3 derivative. This derivative has a modification on the C-terminus that appears to completely block its ability to bind to the receptor. The dissociation constants (K
_d) of the remaining seven ¹¹¹In-labeled peptides are in the nanomolar range and are in agreement with previously published data on this same cell line with similar derivatives [13, 19, 21]. It has been previously reported [11] that the addition of a DPhe residue at the N-terminus would positively affect peptide-binding affinity; however, we did not observe any significant increase in affinity by comparing the DPhe coupled derivatives (BN6, BN7, and BN8) with their respective counterparts not containing DPhe (BN4, BN2, and BN5).

Serum stability was studied in human serum after labeling. Due to its weekly availability in our laboratory (University Hospital Freiburg) in high quantity, we chose ¹⁷⁷Lu for these experiments. There are some clear differences among the compounds tested, indicating that the amino acid substitutions within the binding sequence have important effect on peptide stability. The native BN1 sequence shows very rapid degradation. The fast degradation of the wild-type BN derivative (BN1) is in agreement with the behavior observed for ¹⁷⁷Lu-AMBA [30] in which the same peptide sequence is modified at the N-terminus with a 4-aminobenzoyl spacer and the DO3A chelator.

The BN8 and BN5 derivatives, the two peptides containing the NMeGly¹¹ as well as the Sta¹³ substitutions, show very long plasma half-lives (over 15 days in both cases). The two compounds differ only for the N-terminal DPhe that appears not to affect plasma stability.

Similarly, the BN4/BN6 pair shows fairly long plasma half-life and, also in this case, the addition of the N-terminal DPhe does not appear to have an impact on stability.

The BN2/BN7 pair having the Cha¹³-Nle¹⁴ substitution and again differing only for the N-terminal DPhe shows superimposable plasma half-lives and the rate of degradation is in agreement with previously reported data on the same peptide sequence labeled with ^99mTc using a different approaches [13]. These two peptides, however, show more rapid degradation compared to the BN5/BN8 and BN4/BN6 pairs that contain the Sta¹³-Leu¹⁴ substitutions.

The high affinity binding of the seven of the tested derivatives suggests they are all suitable for in vivo use as GRP-R targeting radiopharmaceuticals. We chose only four for further in vivo studies. The BN1 peptide served as historical reference as it is the one from which all other peptides are derived. Given the clear similarities in the in vitro properties of the BN4/BN6, BN2/BN7, and BN5/BN8 pairs, we chose one from each pair for the in vivo studies. We therefore characterized the DPhe-containing BN7 and BN8 peptides whereas the non-DPhe-containing BN4 was chosen over the BN6 peptide as the latter has been previously characterized in the same animal model [22]. In the interest of limiting the number of animals and given the fact that very low nonspecific retention in both the PC-3 xenografts and the receptor-rich pancreas has been previously reported for similar DOTA coupled peptides [26], we deliberately chose not to perform blocking experiments.

The BN1 peptide showed very rapid blood clearance, very high prolonged uptake in the receptor-rich pancreas, and lower targeting of the PC-3 xenografts that was transient as it diminished over time. These biodistribution properties are very similar to those previously described for ¹⁷⁷Lu-AMBA [30] that also shows prolonged specific retention in the pancreas, transient accumulation in receptor-positive tumor, and rapid blood clearance. The remaining three peptides, BN4, BN7, and BN8, show some similar pattern in biodistribution properties.

All three show transient high accumulation in the PC-3 tumor and pancreas that drops over time. Although tumor accumulation rapidly decreases over time, there is still a fairly wide time window where high target-to-background ratios were observed. The missing DPhe residue in BN4 may be responsible for the poor pharmacokinetic performance of this analog compared to that reported for the DPhe derivative [22]. The uptake in tumor and receptor-positive organs such as the pancreas is reasonably high at the early time point but rapidly decreases over time. The biodistribution profile of the His-Cha derivative (BN7) mirrors the one reported previously on a similar molecule with high uptake in the pancreas at the early time point that decreases over time and with overall fairly low tumor uptake [21]. A favorable pharmacokinetic profile is shown by BN8 with high tumor uptake and good tumor-to-non-target-tissue ratios already at 1 h. This analog is the only one showing the highest uptake in the xenograft compared to all other organs at all times tested. Given this feature and the advantageous target-to-background ratios observed, this derivative seems to have the best characteristics among the eight compounds tested, particularly for imaging applications using short half-lived isotopes such as ⁶⁸Ga.