Research ArticleCancer Imaging

68Ga-DOTA-GGNle-CycMSHhex targets the melanocortin-1 receptor for melanoma imaging

See allHide authors and affiliations

Science Translational Medicine  07 Nov 2018:
Vol. 10, Issue 466, eaau4445
DOI: 10.1126/scitranslmed.aau4445

Visualizing the target

Metastatic melanoma (skin cancer) has a high mortality rate. Yang et al. designed radiolabeled positron emission tomography and fluorescent imaging probes targeting melanocortin-1 receptor, which is overexpressed in melanoma cells. The probes detected melanoma cells in vitro, in xenografts and mouse models, and in metastases in two patients with melanoma. These results suggest that melanocortin-1 receptor-targeting probes may be useful for melanoma imaging.

Abstract

Melanocortin-1 receptor (MC1R) is a molecular target for melanoma imaging and therapy because of its overexpression on rodent and human melanoma cells. Here, we evaluated the MC1R targeting and specificity of 68Ga-DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex using murine and human melanoma cells, and murine and xenografted tumors. 68Ga-DOTA-GGNle-CycMSHhex was used first in human as an imaging probe to evaluate the possibility of radionuclide therapy in patients with advanced-stage melanoma. 68Ga-DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex displayed MC1R-specific targeting properties in murine and human melanoma cells, as well as in murine melanoma and human melanoma–xenografted tumors. Both B16/F10 and M21 melanoma lesions could be easily imaged by positron emission tomography using 68Ga-DOTA-GGNle-CycMSHhex. The first-in-human images of melanoma brain metastases in patients demonstrated the clinical relevance of MC1R as a molecular target for melanoma imaging, highlighting the potential of 68Ga-DOTA-GGNle-CycMSHhex as an MC1R-targeting melanoma imaging probe and underscoring the need to develop MC1R-targeting therapeutic agents for treating patients with metastatic melanoma.

INTRODUCTION

Malignant melanoma is the most lethal form of skin cancer with an increasing rate of incidence. Although malignant melanoma makes up less than 5% of skin cancer cases, it accounts for 75% of all skin cancer deaths (1). About 76,380 new cases and 10,130 fatalities occurred in the United States in 2016 (1). The high mortality rate of melanoma is due to the extreme aggressiveness associated with metastatic melanoma. Unfortunately, the success of traditional treatments, such as dacarbazine chemotherapy, interleukin-2, and interferon-α immunotherapy, is limited for metastatic melanoma, with a traditional median overall survival of 6 to 9 months (2, 3).

Over the past several years, new molecular approaches including targeting BRAF-V600E mutation, cytotoxic T lymphocyte antigen 4 (CTLA-4), and programmed death-1 receptor (PD-1) have shown promising results in treating metastatic melanoma (48). For instance, the treatments of vemurafenib (BRAF inhibitor), ipilimumab (targeting CTLA-4), and nivolumab (PD-1 inhibitor) have improved median overall survival by months in patients with metastatic melanoma (48). However, long-term survival of patients with metastatic melanoma remains at <10%. Therefore, it is highly desirable to develop new treatments for metastatic melanoma.

Melanocortin-1 receptor (MC1R) is a G protein–coupled receptor overexpressed on both rodent and human melanoma cells (916). Greater than 80% of melanotic and amelanotic human metastatic melanomas overexpresses MC1R (9), making it an attractive molecular target for developing receptor-targeting radiopharmaceuticals for melanoma imaging and therapy. Meanwhile, α–melanocyte-stimulating hormone (α-MSH) and its analogs can bind to MC1Rs with nanomolar binding affinities, suggesting the feasibility of using α-MSH peptides to deliver diagnostic and therapeutic radionuclides to melanoma cells via peptide-MC1R binding for melanoma imaging and therapy (1727). Upon MC1R binding, α-MSH peptides can be internalized into melanoma cells (28, 29).

Over several years, we have developed radiolabeled lactam-cyclized α-MSH peptides for melanoma targeting, built upon the construct of DOTA-GGNle-CycMSHhex (1,4,7,10-tetraazacyclononane-1,4,7,10-tetraacetic acid-Gly-Gly-Nle-c[Asp-His-D-Phe-Arg-Trp-Lys]-CONH2) (24, 25, 27, 3034). DOTA-GGNle-CycMSHhex was readily radiolabeled with 111In and 67Ga for melanoma imaging using single photon emission computed tomography (SPECT) (24, 30) and 177Lu for potential targeted radionuclide therapy of melanoma (32). High B16/F1 melanoma uptake (25.53 ± 2.22% ID/g), fast urinary clearance (84.91 ± 2.81% ID), and high tumor/normal organ uptake ratios at 2 hours after injection of 67Ga-DOTA-GGNle-CycMSHhex (30) demonstrated its potential as an imaging probe for melanoma imaging using SPECT. High B16/F1 melanoma uptake (21.63 ± 6.27% ID/g) at 2 hours after injection and prolonged melanoma retention (8.24 ± 1.51% ID/g) at 24 hours after injection of 177Lu-DOTA-GGNle-CycMSHhex (32) highlighted its potential for targeted radionuclide therapy of melanoma.

Building upon the promising results of 67Ga-DOTA-GGNle-CycMSHhex and 177Lu-DOTA-GGNle-CycMSHhex, we developed positron emission tomography (PET) imaging probes for melanoma, taking advantage of PET’s high sensitivity and fast acquisition features. The combination of our MC1R-targeting peptide (DOTA-GGNle-CycMSHhex) with PET may potentially offer a highly sensitive molecular approach to select patients with MC1R-positive melanoma for targeted radionuclide therapy. Gallium-68 [T1/2 (half-life) = 68 min, 89% β+, β+ Emax = 1.899 MeV] is an attractive diagnostic PET radionuclide that can be readily obtained through a commercially available in-house 68Ge/68Ga generator, enabling the production of 68Ga radiopharmaceuticals independent of an on-site cyclotron. The short half-life of 68 min generates minimal radiation doses to patients over time. Meanwhile, the parent radionuclide of 68Ge has a half-life of 270.8 days, making the shelf life of the 68Ge/68Ga generator to be about 6 to 12 months based on elution schedules. As such, the 68Ge/68Ga generator is affordable and offers flexibility to the preparation of 68Ga radiopharmaceuticals.

Although PET imaging using the MC1R-targeting peptide (68Ga-DOTA-GGNle-CycMSHhex) may facilitate the accurate imaging of malignant melanoma, optical fluorescence imaging is a powerful complementary tool to visualize tumor lesions that can potentially provide an opportunity to improve surgical outcome via imaging-guided surgery. Therefore, we developed an MC1R-targeting fluorescence imaging probe in parallel with the MC1R-targeting PET imaging probe (68Ga-DOTA-GGNle-CycMSHhex). By replacing the DOTA chelator with fluorescence Cyanine5.5 (Cy5.5) carboxylic acid and conjugating to GGNle-CycMSHhex, we generated Cy5.5-GGNle-CycMSHhex. The combination of MC1R-targeted PET and fluorescence imaging probes could potentially provide opportunities for accurate detection and imaging-guided surgical removal of melanoma lesions, as well as identification of patients with MC1R-positive melanoma for targeted radionuclide therapy.

Here, DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex were synthesized using 9-fluorenyl methoxycarbonyl (Fmoc) chemistry. Their receptor binding affinities were determined on B16/F10 murine (metastatic) and M21 (low MC1R density) human melanoma cells (16, 17, 34, 35). The MC1R expression in B16/F10 and M21 melanoma cells and tumor lesions was stained by Cy5.5-GGNle-CycMSHhex. The biodistribution and imaging properties of 68Ga-DOTA-GGNle-CycMSHhex were examined on B16/F10 flank melanoma and pulmonary metastatic melanoma models and M21 human melanoma xenografts. 68Ga-DOTA-GGNle-CycMSHhex PET was used first in human on patients with metastatic melanoma who were considered suitable for the evaluation of experimental approaches.

RESULTS

Peptide synthesis and in vitro competitive binding assay

DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex (Fig. 1) were synthesized and purified by reversed-phase high-performance liquid chromatography (RP-HPLC). After purification, DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex displayed greater than 90% purity. The identities of DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex were confirmed by electrospray ionization mass spectrometry. The measured molecular weights matched with calculated molecular weights. The half-maximal inhibitory concentration (IC50) values of DOTA-GGNle-CycMSHhex were 2.1 ± 0.3 nM and 1.1 ± 0.2 nM on B16/F10 and M21 melanoma cells, respectively. The IC50 values of Cy5.5-GGNle-CycMSHhex were 7.1 ± 0.1 nM and 2.6 ± 0.3 nM on B16/F10 and M21 melanoma cells, respectively (Fig. 1).

Fig. 1 Schematic structures and in vitro competitive binding curves.

(A) Schematic structures of DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex. (B) In vitro competitive binding curves of DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex using B16/F10 and M21 melanoma cells. The IC50 values of DOTA-GGNle-CycMSHhex were 2.1 ± 0.3 nM (*, black) and 1.1 ± 0.2 nM (•, pink) on B16/F10 and M21 melanoma cells, respectively. The IC50 values of Cy5.5-GGNle-CycMSHhex were 7.1 ± 0.1 nM (□, green) and 2.6 ± 0.3 nM (○, red) on B16/F10 and M21 melanoma cells, respectively.

Radiochemistry, specific binding, and fluorescence staining

68Ga-DOTA-GGNle-CycMSHhex was readily prepared with greater than 95% radiolabeling yield and was completely separated from excess nonlabeled peptide by RP-HPLC. The retention time of 68Ga-DOTA-GGNle-CycMSHhex and DOTA-GGNle-CycMSHhex was 16.9 and 15.7 min, respectively. 68Ga-DOTA-GGNle-CycMSHhex was stable in mouse serum at 37°C for 1 hour. Urine analysis revealed that 80% of the 68Ga-DOTA-GGNle-CycMSHhex excreted renally also remained intact in urine 2 hours after injection. 68Ga-DOTA-GGNle-CycMSHhex displayed receptor-mediated binding on B16/F10 and M21 cells. About 95 and 84% of 68Ga-DOTA-GGNle-CycMSHhex uptake was blocked on B16/F10 and M21 melanoma cells, respectively (P < 0.05; Fig. 2). The MC1R expression in B16/F10 and M21 cells and in melanoma lesions was examined with fluorescence staining using Cy5.5-GGNle-CycMSHhex. As demonstrated in Fig. 3, Cy5.5-GGNle-CycMSHhex showed receptor-mediated binding on B16/F10 and M21 cells, B16/F10 and M21 flank melanoma lesions, and B16/F10 pulmonary metastatic melanoma lesions. The uptake of Cy5.5-GGNle-CycMSHhex was substantially inhibited by peptide blockade of MC1R.

Fig. 2 UV and radioactive HPLC profiles.

(A) Ultraviolet (UV) HPLC profile of DOTA-GGNle-CycMSHhex (retention time, TR = 15.7 min) and (B) radioactive HPLC profile of 68Ga-DOTA-GGNle-CycMSHhex (TR = 16.9 min). (C) Serum stability of 68Ga-DOTA-GGNle-CycMSHhex at 1 hour after incubation (TR = 16.4 min). (D) Radioactive HPLC profile of the urine sample of a normal C57 mouse at 2 hours after injection of 68Ga-DOTA-GGNle-CycMSHhex (TR = 16.9 min). Arrows indicate the original compounds of DOTA-GGNle-CycMSHhex (A) and 68Ga-DOTA-GGNle-CycMSHhex (B to D). Specific binding of 68Ga-DOTA-GGNle-CycMSHhex on (E) B16/F10 and (F) M21 melanoma cells with or without peptide blockade (P < 0.05).

Fig. 3 Fluorescence staining of MC1R.

Fluorescence staining of (A) B16/F10 murine melanoma cells, (B) B16/F10 flank melanoma lesions, (C) B16/F10 pulmonary metastatic melanoma lesions, (D) M21 human melanoma cells, and (E) M21-xenografted flank melanoma lesions. Cells and melanoma tissues were stained with 1 μM Cy5.5-GGNle-CycMSHhex (red) in the presence or absence of 100 μM peptide blockade. The nuclei were stained with DAPI (blue). The microscopic images were acquired by confocal laser microscopy at 100× magnification. Scale bar, 20 μm.

Biodistribution

Tumor targeting and biodistribution properties of 68Ga-DOTA-GGNle-CycMSHhex were determined in B16/F10 flank melanoma–bearing C57 mice. The biodistribution results of 68Ga-DOTA-GGNle-CycMSHhex are presented in Table 1. 68Ga-DOTA-GGNle-CycMSHhex displayed rapid melanoma uptake. The B16/F10 tumor uptake was 18.63 ± 4.97 and 24.27 ± 3.74% ID/g at 0.5 and 1 hour after injection, respectively. 68Ga-DOTA-GGNle-CycMSHhex exhibited prolonged tumor retention, with 11.61 ± 2.87% ID/g of tumor uptake at 2 hours after injection. The co-injection of nonradioactive [Nle4, D-Phe7]-α-melanocyte-stimulating hormone (NDP-MSH) blocked 92% of the tumor uptake at 1 hour after injection, demonstrating that the tumor uptake was MC1R mediated. Whole-body clearance of 68Ga-DOTA-GGNle-CycMSHhex was rapid, with 85% of the injected dose being washed out of the body via the renal system by 1 hour after injection and 92% of the injected dose cleared by 2 hours after injection. Kidneys have the highest uptake of 68Ga-DOTA-GGNle-CycMSHhex of normal organs, and renal uptake was 9.22 ± 1.38, 7.78 ± 0.63, and 6.01 ± 0.91% ID/g at 0.5, 1, and 2 hours after injection, respectively. The co-injection of NDP-MSH did not significantly (P > 0.05) reduce the renal uptake, indicating that the renal uptake of 68Ga-DOTA-GGNle-CycMSHhex was not receptor mediated. The accumulation of 68Ga-DOTA-GGNle-CycMSHhex in other normal organs was much lower than that in kidneys. High tumor/blood and tumor/normal organ uptake ratios were demonstrated as early as 0.5 hour after injection.

Table 1 Biodistribution of 68Ga-DOTA-GGNle-CycMSHhex on B16/F10 murine melanoma–bearing C57 mice.

The data are presented as percent injected dose per gram (%ID/g) or as percent injected dose (%ID) (means ± SD, n = 4).

View this table:

The tumor targeting and biodistribution properties of 68Ga-DOTA-GGNle-CycMSHhex were also determined in M21 human melanoma–xenografted nude mice (Table 2). As compared to the biodistribution results in B16/F10 melanoma–bearing mice, 68Ga-DOTA-GGNle-CycMSHhex displayed a similar distribution pattern with fast tumor uptake, prolonged tumor retention, and quick urinary clearance in M21 melanoma–xenografted mice. The M21 tumor uptake was 6.05 ± 0.83, 6.07 ± 0.68, and 5.15 ± 0.78 % ID/g at 0.5, 1, and 2 hours after injection, respectively. The tumor uptake was MC1R specific because the co-injection of nonradioactive NDP-MSH blocked 83% of the tumor uptake at 1 hour after injection. M21 tumor uptake was lower than B16/F10 tumor uptake, which is likely attributed to lower MC1R receptor density (34) and to differences in tumor morphology. The M21 tumor is solid unlike the soft and vascularized B16/F10 tumor. Kidneys exhibit the highest uptake of 68Ga-DOTA-GGNle-CycMSHhex of all normal organs; however, the renal uptake of 68Ga-DOTA-GGNle-CycMSHhex was not MC1R specific because the co-injection of NDP-MSH did not significantly (P > 0.05) reduce renal uptake. Similarly, the accumulation of 68Ga-DOTA-GGNle-CycMSHhex in other normal organs was much lower than that in the kidneys. Thus, 68Ga-DOTA-GGNle-CycMSHhex reached high tumor/blood and tumor/normal organ uptake ratios as early as 0.5 hour after injection.

Table 2 Biodistribution of 68Ga-DOTA-GGNle-CycMSHhex on M21 human melanoma–xenografted nude mice.

The data are presented as percent injected dose/gram (%ID/g) or as percent injected dose (%ID) (means ± SD, n = 4).

View this table:

Melanoma imaging in tumor-bearing mice

Both flank B16/F10 tumors and M21-xenografted tumors could be visualized by PET using 68Ga-DOTA-GGNle-CycMSHhex as an imaging probe at 0.5, 1, and 2 hours after injection (Fig. 4). 68Ga-DOTA-GGNle-CycMSHhex displayed the best imaging contrast between tumor and other normal organs at 1 hour after injection for both B16/F10 and M21 flank tumor models, consistent with the biodistribution results. In the B16/F10 pulmonary metastatic melanoma model, 68Ga-DOTA-GGNle-CycMSHhex exhibited substantial uptake in the melanoma-bearing lung instead of sharp images of metastatic melanoma lesions (Fig. 5).

Fig. 4 PET imaging of flank melanoma.

Representative PET imaging of a live B16/F10 murine melanoma–bearing mouse (top) and a live M21 human melanoma–xenografted nude mouse (bottom) using 68Ga-DOTA-GGNle-CycMSHhex as an imaging probe at 0.5, 1, and 2 hours after injection. Melanoma lesions are highlighted with arrows on the images.

Fig. 5 PET imaging of pulmonary metastatic melanoma.

Representative PET imaging of a live B16/F10 pulmonary metastatic melanoma–bearing mouse using 68Ga-DOTA-GGNle-CycMSHhex as an imaging probe at 0.5 and 1 hour after injection. Pulmonary metastatic melanoma lesions are highlighted with arrows on the images.

First-in-human melanoma imaging

Similar to preclinical studies, first-in-human imaging of two patients with melanoma demonstrated moderate to strong uptake in metastatic melanoma lesions (Fig. 6). Three melanoma metastases in brain, one metastasis in lung, one metastasis in connective tissue, and bulky metastases in small intestine of patient 1 (Fig. 6, A and C) were easily visualized by PET using 68Ga-DOTA-GGNle-CycMSHhex. The routine staging per 18F-FDG PET/CT (Fig. 6, B and D) was non-inferior and therefore also confirmative for true-positive findings regarding the detection of extracerebral lesions. However, because of the high physiological uptake of 18F-FDG in the brain, additional magnetic resonance imaging (MRI; BM1 and BM2 in Fig. 6) was performed to confirm the brain metastases using conventional means. For patient 2, massive melanoma metastases on the left leg and groin lymph node were visualized by PET using either 68Ga-DOTA-GGNle-CycMSHhex or 18F-FDG (Fig. 6, E and F). As compared to the more homogeneous uptake of 18F-FDG in those metastases, the uptake of 68Ga-DOTA-GGNle-CycMSHhex was more heterogeneous, suggesting different MC1R expressions among those metastases.

Fig. 6 PET imaging of patients with melanoma.

Representative first-in-human PET studies on two patients with melanoma. Melanoma metastases in the brain, lung, connective tissue, and small intestine of patient 1 (A and C) were visualized by PET using 68Ga-GGNle-CycMSHhex. Melanoma metastases on the left leg of patient 2 (E) were imaged by PET using 68Ga-GGNle-CycMSHhex. 18F-FDG PET (B, D, and F) studies were performed on the same patients for comparisons. The melanoma metastases in brain (BM1 and BM2) were confirmed by MRI, and the metastasis in connective tissue (CM1) was confirmed by CT. Melanoma metastases are highlighted with arrows on the images. SUVbw, standardized uptake value based on body weight.

DISCUSSION

Malignant melanoma is the most lethal form of skin cancer owing to the extreme aggressiveness associated with metastasis. We have been developing radiolabeled α-MSH peptides to target MC1Rs for melanoma imaging and therapy because MC1R is expressed on more than 80% of melanotic and amelanotic human melanoma metastases (9). Here, we have demonstrated MC1R targeting and specificity of 68Ga-DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex using murine and human melanoma cells, as well as murine melanoma and human melanoma–xenografted tumors. Both B16/F10 and M21 melanoma lesions could be easily imaged by PET using 68Ga-DOTA-GGNle-CycMSHhex. MC1R expression in B16/F10 and M21 melanoma lesions could be specifically stained by Cy5.5-GGNle-CycMSHhex. The promising results on MC1R-specific imaging and staining on both flank melanoma and pulmonary metastatic melanoma lesions suggested the potential of using Cy5.5-GGNle-CycMSHhex for imaging-guided surgery after identifying melanoma lesions using 68Ga-DOTA-GGNle-CycMSHhex. The combined use of 68Ga-DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex may potentially improve the surgical outcomes of patients with melanoma via imaging-guided surgery.

We and others have been using the B16/F10 pulmonary metastatic melanoma model to evaluate the targeting and imaging properties of radiolabeled α-MSH peptides (16, 17, 34) due to the high metastatic nature of the B16/F10 cell line. B16/F10 melanoma cells were injected into mice via tail veins to generate pulmonary melanoma metastases. This inoculation route mimics the development of melanoma metastases through blood circulation, whereas sentinel lymph node is the first organ to develop metastasis in a patient with melanoma. Both B16/F10 flank melanoma lesions and B16/F10 pulmonary metastatic melanoma lesions were positively stained by Cy5.5-GGNle-CycMSHhex, indicating that the metastasis process did not result in a substantial change in the MC1R expression. PET images showed substantial uptake of 68Ga-DOTA-GGNle-CycMSHhex in B16/F10 pulmonary metastatic melanoma lesions. In our previous work, we demonstrated the success of SPECT imaging of B16/F10 pulmonary melanoma metastases using 99mTc- and 67Ga-labeled lactam-cyclized α-MSH peptides as imaging probes (34, 36). The difference in imaging quality between SPECT and PET images in this B16/F10 pulmonary metastatic melanoma model was likely attributed to the lower resolution of animal PET as compared to animal SPECT.

After diagnosis of melanoma, it is important to determine whether the melanoma has metastasized to lymph nodes and other distant organs such as liver, lung, and brain. About 60% of patients with metastatic melanoma develop brain metastases during the course of their disease (37). Patients with brain metastases have much shorter life expectancies than patients without brain metastases. Meanwhile, it is challenging to effectively treat brain metastases without side effects. Here, we could identify the melanoma metastases in brain, as well as in lung, connective tissue, and small intestine of patients by PET using 68Ga-DOTA-GGNle-CycMSHhex. The sharp images of these metastases in patients demonstrated the clinical relevance of MC1R as a valid and attractive molecular target for melanoma imaging, highlighting the potential of 68Ga-DOTA-GGNle-CycMSHhex as an MC1R-targeting probe for melanoma imaging. This underscores the need to develop MC1R-targeting therapeutic agents for treating patients with metastatic melanoma.

The PET images of melanoma metastases suggest that MC1R not only is a valid molecular target for melanoma during early development but also is present in advanced stages of melanoma and later generations of daughter metastases. Moreover, the heterogeneous intensity of 68Ga-DOTA-GGNle-CycMSHhex uptake among melanoma metastases in patient 2 indicates that the MC1R density was different in these metastatic melanoma lesions. From a therapeutic point of view, it is important to note that receptor-targeted radionuclide therapy is much more effective when the receptors are overexpressed in all or at least nearly all tumor lesions. Therefore, it is crucial to identify such patients for potentially successful receptor-targeted radionuclide therapy. The PET images of patients with melanoma demonstrate the potential of 68Ga-DOTA-GGNle-CycMSHhex PET to select patients with sufficient MC1R expression in all metastases before tailoring therapies for these patients toward or against MC1R-targeted radionuclide therapy. We have previously demonstrated the promising melanoma targeting property of 177Lu-DOTA-GGNle-CycMSHhex (32). 177Lu is an attractive theranostic radionuclide. When 177Lu decays, it generates medium-energy (0.497 MeV) β-particles for treatment, as well as γ-rays (113 and 208 keV) for SPECT imaging. The short tissue penetration of β-particles makes 177Lu a suitable radionuclide for treating small tumors and metastases because of its fairly localized radiation emission. Although we have yet to examine the therapeutic efficacy of 177Lu-DOTA-GGNle-CycMSHhex, the combination of 68Ga-DOTA-GGNle-CycMSHhex and 177Lu-DOTA-GGNle-CycMSHhex may potentially offer an attractive approach to receptor-targeted imaging and radionuclide therapy for melanoma.

MC1R is a melanoma-specific molecular target without relevant physiological expression on normal tissues. Despite DOTA-GGNle-CycMSHhex exhibiting low nanomolar affinity to MC1R, the relatively low number of MC1R copies per cell could be a potential limiting factor for MC1R-targeted radionuclide therapy. The human imaging doses of 68Ga-DOTA-GGNle-CycMSHhex were prepared with 7.5 μg of DOTA-GGNle-CycMSHhex in this study. However, it is possible that higher amounts of GGNle-CycMSHhex would be needed to deliver more 177Lu activity to melanoma cells for MC1R-targted radionuclide therapy. Therefore, there is a potential scenario in which increased GGNle-CycMSHhex mass could partially saturate MC1Rs on melanoma cells. More research is needed to determine the threshold amount of GGNle-CycMSHhex without MC1R blocking effect and to achieve high specific activity during radiolabeling or through an additional purification process. Nonspecific renal uptake of 68Ga-DOTA-GGNle-CycMSHhex could be another potential limiting factor for MC1R-targeted radionuclide therapy. Nonspecific renal uptake of radiolabeled peptide is due to the electrostatic interaction between positively charged peptide molecules and the negatively charged surface of tubule cells as the peptides are filtered in the glomerulus and reabsorbed in the cells of the proximal tubule (38). Positively charged lysine has been effective in reducing the renal uptake of radiolabeled peptide by shielding such electrostatic interaction (3941), suggesting that co-injection of lysine could be used to further reduce renal uptake if needed.

In conclusion, we demonstrated the MC1R targeting and specificity of 68Ga-DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex using murine and human melanoma cells, as well as murine melanoma and human melanoma–xenografted tumors. Both B16/F10 and M21 melanoma lesions could be imaged by PET using 68Ga-DOTA-GGNle-CycMSHhex as an imaging probe. The combination of 68Ga-DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex may potentially improve surgical outcome via imaging-guided surgery. The first-in-human images of melanoma metastases in patients demonstrated the clinical relevance of MC1R as a valid and attractive molecular target for melanoma imaging, as well as highlighted the potential of 68Ga-DOTA-GGNle-CycMSHhex as an MC1R-targeting melanoma imaging probe. Furthermore, the successful imaging of melanoma metastases in patients underscored the need to develop MC1R-targeting therapeutic peptides for treating patients with metastatic melanoma.

MATERIALS AND METHODS

Study design

The goal of this work was to assess whether MC1R-targeting 68Ga-DOTA-GGNle-CycMSHhex could be used to evaluate the possibility of radionuclide therapy on patients with advanced-stage melanoma. First, the MC1R targeting and specificity of 68Ga-DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex were examined using murine and human melanoma cells, as well as murine melanoma and human melanoma–xenografted tumors. Then, 68Ga-DOTA-GGNle-CycMSHhex was used first in human as an imaging probe to evaluate the possibility of radionuclide therapy on patients with metastatic melanoma who were considered suitable for the evaluation of experimental approaches (in agreement with the Declaration of Helsinki, 37: Unproven Interventions in Clinical Practice) because they had already exhausted all other available treatment options.

Chemicals and reagents

Amino acids and resin were purchased from Advanced ChemTech Inc. and Novabiochem. DOTA-tris-t-butyl ester and Cy5.5 carboxylic acid were purchased from Macrocyclics Inc. and Lumiprobe Corporation, respectively, for peptide synthesis. The 68Ge-68Ga generator was purchased from RadioMedix Inc. All other chemicals used in this study were purchased from Thermo Fisher Scientific and used without further purification. Four percent paraformaldehyde (PFA) in phosphate-buffered saline (PBS) was obtained from Alfa Aesar, xylene was obtained from Fisher Chemical, and 4′,6-diamidino-2-phenylindole (DAPI) Fluoromount-G mounting medium was obtained from SouthernBiotech. ProLong Diamond antifade mounting reagent with DAPI was obtained from Life Technologies. B16/F10 murine melanoma cells were obtained from the American Type Culture Collection. M21 human melanoma cells were supplied by D. A. Cheresh from the Department of Pathology, University of California San Diego Moores Cancer Center.

Peptide synthesis

DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex were synthesized using standard Fmoc chemistry according to our published procedure (23) with modifications. Briefly, linear peptide backbones of (tBu)3DOTA-Gly-Gly-Nle-Asp(O-2-PhiPr)-His(Trt)-DPhe-Arg(Pbf)-Trp(Boc)-Lys(Dde) and Cy5.5-Gly-Gly-Nle-Asp(O-2-PhiPr)-His(Trt)-DPhe-Arg(Pbf)-Trp(Boc)-Lys(Dde) were synthesized on Sieber amide resin by an Advanced ChemTech multiple-peptide synthesizer. Seventy micromoles of resin, 210 μmol of each Fmoc-protected amino acid, 210 μmol of (tBu)3DOTA, and Cy5.5 were used for the synthesis. The protecting group of Dde was removed by 2% hydrazine for peptide cyclization. The protecting group of 2-phenylisopropyl was removed, and the protected peptide was cleaved from the resin treating with a mixture of 2.5% of trifluoroacetic acid (TFA) and 5% of triisopropylsilane. After the precipitation with ice-cold ether, each protected peptide was characterized by liquid chromatography–mass spectroscopy (LC-MS). Then, each protected peptide was further cyclized by coupling the carboxylic group from the Asp with the ε-amino group from the Lys. The cyclization reaction was achieved by an overnight reaction in dimethylformamide using benzotriazole-1-yl-oxy-tris-pyrrolidino-phosphonium-hexafluorophosphate as a coupling agent in the presence of N,N-diisopropylethylamine. After characterization by LC-MS, each cyclized protected peptide was dissolved in H2O/CH3CN (50:50) and lyophilized. The protecting groups were totally removed by treatment with a mixture of TFA, thioanisole, phenol, water, ethanedithiol, and triisopropylsilane (87.5:2.5:2.5:2.5:2.5:2.5) for 2 hours at room temperature (25°C). Each peptide was precipitated and washed four times with ice-cold ether, purified by RP-HPLC, and characterized by LC-MS.

In vitro competitive binding assay of DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex

The MC1R binding affinities of DOTA-GGNle-CycMSHhex and Cy5.5-GGNle-CycMSHhex were determined on B16/F10 and M21 melanoma cells by in vitro competitive receptor binding assay. The receptor binding assay was replicated in triplicate. B16/F10 and M21 cells (2 × 105 cells per well, n = 3) were seeded separately in 24-well cell culture plates and incubated overnight at 37°C. After being washed with binding medium [Dulbecco’ s modified Eagle’s medium with 25 mM N-(2-hydroxyethyl)-piperazine-N′-(2-ethanesulfonic acid) (pH 7.4), 0.2% bovine serum albumin (BSA), and 0.3 mM 1,10-phenanthroline], the cells were incubated at 37°C for 2 hours with about 45,000 cpm of 125I-(Tyr2)-NDP-MSH in the presence of 10−12 to 10−5 M of the peptide in 0.3 ml of binding medium. The binding medium was aspirated after the incubation. The cells were rinsed twice with 0.5 ml of ice-cold 0.2% BSA/0.01 M PBS (pH 7.4) and lysed in 0.5 ml of 1 N NaOH for 5 min. The cells were harvested and measured in a PerkinElmer Wallac 1480 automated γ-counter. The IC50 value was calculated using GraphPad Prism software.

Preparation, serum stability, and urinary metabolites of 68Ga-DOTA-GGNle-CycMSHhex

The 68Ge-68Ga generator was eluted with 4 ml of 0.05 M HCl via fractional elutions (1 ml per fraction) to obtain 68GaCl3. 68Ga-DOTA-GGNle-CycMSHhex was prepared in a 2 M NH4OAc-buffered solution (pH 5). Briefly, 300 μl of 68GaCl3 [37 to 74 megabecquerel (MBq) in 0.05 M HCl aqueous solution], 20 μl of peptide aqueous solution (1 mg/ml), and 40 μl of 2 M NH4OAc (pH 5) were added into a reaction vial and incubated at 95°C for 5 min. The pH of the reaction mixture was 4.5. After the incubation, 10 μl of 0.5% EDTA aqueous solution was added into the reaction vial to scavenge potentially unbound 68Ga3+ ions. The radiolabeled complexes were purified to single species by a Waters RP-HPLC on a Grace Vydac C-18 reversed-phase analytical column using the following gradient at a flow rate of 1 ml/min. The mobile phase consisted of solvent A (20 mM HCl aqueous solution) and solvent B (100% CH3CN). The gradient was initiated and kept at 82:18 A/B for 3 min, followed by a linear gradient of 82:18 A/B to 72:28 A/B over 20 min. Then, the gradient was changed from 72:28 A/B to 10:90 A/B over 3 min, followed by an additional 5 min at 10:90 A/B. Thereafter, the gradient was changed from 10:90 A/B to 82:18 A/B over 3 min. The purified peptide sample was purged with N2 gas for 15 min to remove the acetonitrile. The pH of the final solution was adjusted to 7.4 with 0.1 N NaOH and sterile saline for animal studies. In vitro serum stability of 68Ga-DOTA-GGNle-CycMSHhex was determined by incubation in mouse serum at 37°C for 1 hour and monitored for degradation by RP-HPLC. To determine urine metabolites, HPLC-purified 68Ga-DOTA-GGNle-CycMSHhex (0.74 MBq, 100 μl) was injected into a normal C57 mouse through the tail vein. At 2 hours after injection, the mouse was euthanized, and the urine was collected. The radioactive metabolites in the urine were analyzed by HPLC. A 20-min gradient of 18 to 28% CH3CN/20 mM HCl was used for the urine analysis.

Specific binding of 68Ga-DOTA-GGNle-CycMSHhex on B16/F10 and M21 melanoma cells

The specific binding of 68Ga-DOTA-GGNle-CycMSHhex was determined using B16/F10 and M21 melanoma cells. The B16/F10 and M21 cells (1 × 106 cells per tube, n = 3) were incubated at 37°C for 2 hours with about 0.037 MBq of 68Ga-DOTA-GGNle-CycMSHhex with or without 10 μg (6.07 nmol) of unlabeled NDP-MSH in 0.3 ml of binding medium [Dulbecco’s modified Eagle’s medium with 25 mM N-(2-hydroxyethyl)-piperazine-N′-(2-ethanesulfonic acid) (pH 7.4), 0.2% BSA, and 0.3 mM 1,10-phenanthroline]. The binding medium was aspirated after the incubation. The cells were rinsed three times with 0.5 ml of ice-cold 0.2% BSA/0.01 M PBS (pH 7.4) and measured in a PerkinElmer Wallac 1480 automated γ-counter.

B16/F10 murine melanoma and M21 human melanoma–xenografted melanoma models

All animal studies were conducted in compliance with Institutional Animal Care and Use Committee approval. B16/F10 murine melanoma and M21 human melanoma–xenografted melanoma models were generated for fluorescence staining, biodistribution studies, and imaging studies. For the B16/F10 flank melanoma model, each C57 mouse (4 to 6 weeks, female; Charles River Laboratories) was subcutaneously inoculated with 1 × 106 B16/F10 cells on the right flank. Ten days after inoculation, tumor weights reached about 0.2 g. The flank tumor–bearing mice were used for fluorescence staining, biodistribution studies, and imaging studies. For the B16/F10 pulmonary metastatic melanoma model, each C57 mouse was intravenously injected with 2 × 105 B16/F10 cells via the tail vein. The mice were used for fluorescence staining 16 days after injection and for imaging studies 21 days after injection. For M21 human melanoma xenografts, each nude mouse (4 to 6 weeks, female; Charles River Laboratories) was subcutaneously inoculated with 5 × 106 M21 cells on the right flank. Fourteen days after inoculation, the tumor weights reached about 0.2 g, and these mice were used for fluorescence staining, biodistribution studies, and imaging studies.

Fluorescence staining of melanoma cells and tumor lesions using Cy5.5-GGNle-CycMSHhex

B16/F10 and M21 melanoma cells (1 × 105 cells per well) were seeded in a four-well Lab-Tek Chamber Glass Slide System from Thermo Fisher Scientific and incubated overnight at 37°C. After 24 hours, the cells were fixed with 4% PFA in PBS and incubated at room temperature for 15 min, washed with PBS three times, treated with 0.5% Triton X-100 at room temperature for 15 min, and washed with PBS three times. The cells were incubated with 1 μM Cy5.5-GGNle-CycMSHhex with or without 100 μM NDP-MSH peptide blockade at room temperature for 1 hour and washed with PBS three times. Then, the cells were stained for nuclei and mounted with DAPI Fluoromount-G mounting medium from SouthernBiotech and stayed in the dark at room temperature for 24 hours. The fluorescent signal was observed and recorded at 100× magnification under an Olympus FV1000 confocal microscope.

Paraffin-embedded B16/F10 and M21 melanoma sections (thickness, 5 μm) were incubated with 1 μM Cy5.5-GGNle-CycMSHhex with or without 100 μM NDP-MSH peptide blockade at room temperature for 1 hour after deparaffinization with xylene. Tissue samples were then washed with PBS three times and stained and mounted with ProLong Diamond antifade mounting reagent with DAPI from Life Technologies. The fluorescent signal was observed and recorded at 100× magnification under an Olympus FV1000 confocal microscope.

Biodistribution of 68Ga-DOTA-GGNle-CycMSHhex

All animal studies were conducted in compliance with Institutional Animal Care and Use Committee approval. The biodistribution properties of 68Ga-DOTA-GGNle-CycMSHhex were determined on B16/F10 murine melanoma–bearing C57 mice and M21 human melanoma–xenografted nude mice. Each melanoma-bearing mouse was injected with 0.37 MBq of 68Ga-DOTA-GGNle-CycMSHhex via the tail vein. Groups of four mice were euthanized at 0.5, 1, and 2 hours after injection, and tumors and organs of interest were harvested, weighed, and counted. Blood values were taken as 6.5% of the whole-body weight. The specificity of the tumor uptake of 68Ga-DOTA-GGNle-CycMSHhex was determined by co-injecting 10 μg (6.07 nmol) of unlabeled NDP-MSH, which is a linear α-MSH peptide analog with subnanomolar MC1R binding affinity.

PET imaging of 68Ga-DOTA-GGNle-CycMSHhex in melanoma-bearing mice

The melanoma imaging property of 68Ga-DOTA-GGNle-CycMSHhex was determined on B16/F10 flank melanoma–bearing and B16/F10 pulmonary metastatic melanoma–bearing C57 mice and on M21 human melanoma–xenografted nude mice. The melanoma-bearing mice were generated as described above and used for imaging studies. The B16/F10 pulmonary melanoma metastases were generated by injecting B16/F10 cells into the tail vein of each C57 mouse, mimicking the development of metastases via blood circulation. Each melanoma-bearing mouse was injected with 7.4 MBq of 68Ga-DOTA-GGNle-CycMSHhex via the tail vein. PET imaging studies of live flank melanoma–bearing mice were performed at 0.5, 1, and 2 hours after injection, whereas the imaging study of live pulmonary metastatic melanoma–bearing mice was conducted at 0.5 and 1 hour after injection. Reconstructed PET data were visualized using VivoQuant from Invicro.

First-in-human PET imaging of 68Ga-DOTA-GGNle-CycMSHhex in patients with metastatic melanoma

To assess the eventual possibility of MC1R-targeting radionuclide therapy, 68Ga-DOTA-GGNle-CycMSHhex PET/CT was offered to two patients with metastatic melanoma. These patients had already failed other available melanoma treatments including chemotherapy (dacarbazine and paclitaxel/carboplatin), immunotherapy (ipilimumab), and BRAF inhibitor (vemurafenib). PD-1 therapy was not available yet at the time of the study. After exhausting all other available treatment options, it was well in line with the updated Declaration of Helsinki (37: Unproven Interventions in Clinical Practice) to consider nonstandard approaches for these individual patients. The patients were informed about the experimental character of the procedure that was offered according to the German Pharmaceutical Act §13(2b) and were given written informed consent. The ethics committee of the University of Heidelberg approved the retrospective evaluation of the obtained data as an observational study.

68Ga-DOTA-GGNle-CycMSHhex was prepared at pH 3.5 in a 2.5 M NaOAc-buffered solution (pH 9). Briefly, 1 ml of 68GaCl3 (600 to 1000 MBq in 0.6 M HCl aqueous solution), 5 μl of 1 mM peptide aqueous solution, 10 μl of 20% ascorbic acid aqueous solution, and 300 μl of 2.5 M NaOAc were added into a reaction vial and incubated at 95°C for 15 min. The pH of the reaction mixture was 3.5. After the incubation, the mixture was purified by solid-phase extraction (Varian Bond Elut Plexa from Agilent Technologies) before injection. About 300 to 600 MBq of the purified 68Ga-DOTA-GGNle-CycMSHhex was obtained in 6 ml of 10% ethanol and saline.

The patients underwent imaging on a Siemens Biograph 6 scanner 1 hour after intravenous injection of about 300 MBq of 68Ga-DOTA-GGNle-CycMSHhex. First, a non–contrast-enhanced low-dose CT scan (130 keV, 30 mA, and reconstruction with soft tissue kernel to a slice thickness of 5 mm) was performed and later used for attenuation correction of the related PET scan. Then, three-dimensional emission scans from head to mid-thigh were obtained by eight bed positions (15.5 cm field of view and 3 min of acquisition time, respectively). Images were reconstructed with an ordered subset expectation maximization algorithm correcting for randoms, scatter, and decay using two iterations/16 subsets and Gauss filtered to a transaxial resolution of 5 mm at full width at half maximum.

Statistical analysis

Statistical analysis was performed using the Student’s t test for unpaired data. A 95% confidence level was chosen to determine the significant difference between groups in cellular binding of 68Ga-DOTA-GGNle-CycMSHhex and difference in tumor and renal uptakes between 68Ga-DOTA-GGNle-CycMSHhex with or without NDP-MSH co-injection. The differences at the 95% confidence level (P < 0.05) were considered significant.

REFERENCES AND NOTES

Acknowledgments: We thank F. Gallazzi, L. Cheuy, N. Serkova, and K. Huber for technical assistance and E. R. Prossnitz for helpful discussions. Funding: This work was supported in part by NIH grant number R01CA225837 and the University of Colorado Denver start-up fund. Microscopy imaging experiments were performed in the University of Colorado Anschutz Medical Campus Advance Light Microscopy Core supported in part by NIH/NCATS Colorado CTSI grant number UL1 TR001082. PET imaging experiments were conducted in the University of Colorado Anschutz Medical Campus Animal Imaging Shared Resources supported in part by the University of Colorado Cancer Center (NCI P30 CA046934) and the Colorado Clinical and Translational Sciences Institute (NIH/NCATS UL1 TR001082). Author contributions: J.Y. and J.X. were responsible for the execution of experiments, data collection, and statistical analysis. J.Y. also contributed to manuscript preparation. R.G. contributed to research project organization and execution and review and critique of the manuscript. T.L. contributed to human doses production. C.K. was responsible for first-in-human studies on patients with melanoma, image data interpretation, and review and critique of the manuscript. Y.M. was responsible for research project conception, organization, and execution; contributed to the statistical analysis design and execution; wrote the first draft of the manuscript; and reviewed and edited the final draft of the manuscript. Competing interests: Y.M. is an inventor on awarded U.S. patents (US-8,986,651-B2, US-9,493,537-B2, and US-10,047,135-B2) held by the University of New Mexico that cover compounds with reduced ring size for use in diagnosing and treating melanoma, including metastatic melanoma and methods related to same. The other authors declare that they have no competing interests. Data and materials availability: All data associated with this study are present in the paper. Request for peptides will be handled by the corresponding authors through a licensing agreement (for commercial purposes) or a material transfer agreement (for research and noncommercial purposes).
View Abstract

Navigate This Article