Novel Urokinase-Type Plasminogen Activator Receptor-Targeting Optical Imaging Agent ICG-Glu-Glu-AE105 for Visualization of Malignant Glioma During Surgery: First-in-Human Study in 35 Patients with Brain Cancer

Abstract

Background and objectives: Glioblastoma is an aggressive form of brain cancer for which surgery is the keystone in treatment before oncological treatment. Improving surgical resection without jeopardizing the outcome is eminent, and a fluorescent drug to aid surgical outcome is warranted. To evaluate the safety and the efficacy of a novel urokinase-type Plasminogen Activator Receptor-targeting near-infrared optical imaging agent, ICG-Glu-Glu-AE105 (FG001), in patients with malignant glioma (glioblastoma).

Methods: First-in-human phase I dose escalation and time elaboration study in 35 patients undergoing surgery for glioblastoma. The tumor-to-background ratio (TBR) was measured on near-infrared images as an objective measure of image contrast. Biopsies were taken during surgery from areas with and without fluorescence (FG001) and compared with pathology as reference. Efficacy was evaluated as sensitivity and specificity of FG001 to detect tumor tissue. The study was conducted with close safety monitoring.

Results: Administration of FG001 at a dose of 36 mg, 12 to 17 hours before surgery, resulted in optimal image contrast between tumor and normal brain tissue, with a mean TBR of 3.6. A high sensitivity (79%) and specificity (100%) for the detection of tumor tissue was found. Safety monitoring during the study identified only a few related adverse events, all of which were of mild grade.

Conclusion: FG001 was found to be safe and well tolerated in the patients included in the study. The optimal dose of FG001 was selected on the basis of videos taken during the surgery and the measured TBR values. A dose of 36 mg administered 16 hours (mean) before the surgery showed the best contrast (TBR values) and optimal visualization of the tumor delineation. Histology demonstrated good sensitivity of FG001.

Keywords: Brain cancer; Fluorescence; High-grade glioma; Optical guided surgery; Optical imaging; Surgery; uPAR.

Graphical abstract

FIGURE 1.

Overview of trial design shows the flow for eligibility, dosing of experimental drug, patient monitoring, and confirmation of efficacy. FG001, ICG-Glu-Glu-AE105.

FIGURE 2.

Consort flow of patient inclusion. Of the 104 screened patients, 40 met the inclusion and exclusion criteria to be included for surgery after informed consent. Of these, final histology rapport showed 4 patients not to have a glioblastoma and 1 patient withdrew consent, leaving a total of 35 patient with diffuse astrocytoma, IDHwildtype for further analysis.

FIGURE 3.

Images (Zeiss Pentero) show fluorescent signal in 2 patients receiving 36 mg FG001 at different timepoint: A, 3.5 hours before surgery. B, 16.5 hours before surgery. C-E, Show white light, near-infrared and fused images of a patient receiving 36 FG001 at 16.5 hours before surgery. FG001, ICG-Glu-Glu-AE105.

FIGURE 4.

TBR (contrast) and fluorescence intensities in the 36 and 48 mg evening dose cohorts obtained by ZEISS microscope. Because no further increase in TBR was seen from 36 mg, 36 and 48 mg were considered equal in performance. TBR calculation was performed using the image software ImageJ (National Institutes of Health, MD). On the gray scale image, ROIs were drawn on the same frame over the tumor area and over a representative background area, respectively. Intensity is expressed as the value from the gray scale image ranking from “0” (black) to “255” (white). The TBR was calculated as the ratio between the top 10% pixels intensity in the tumor ROI and the mean intensity in the background ROI. FG001, ICG-Glu-Glu-AE105; ROI, regions of interest; TBR, tumor-to-background ratio.

FIGURE 5.

All biopsies were sent to the Department of Pathology for processing. The specimens were fixed in 4% formalin and embedded in paraffin according to standard protocols. Sections (5 µm thick) were cut and stained with hematoxylin and eosin. uPAR staining of biopsies: From the biopsies a 4 μm slice was sectioned. Slides were deparaffinized in Histoclear solution, rehydrated in graded ethanol, ranging from 99% to 70% and ending in water. Antigen retrieval was performed by incubating the slides in pH 6 buffer in the microwave for 10 min. For staining the anti-uPAR antibody GTX100476 (Genetex) with a dilution of 1:500 was incubated on the slides overnight. Secondary staining was performed with horseradice peroxide conjugated antibody, after diamonobenzidine solution. A, Tumor area showing high expression of uPAR. B, Normal brain parenchyma showing no expression of uPAR. C, Image with magnification of labeled area in B showing normal brain parenchyma with neurons. uPAR, urokinase-type Plasminogen Activator Receptor.

FIGURE 6.

The pharmacokinetics profiles for FG001 at the 8 dose levels examined in the phase I trial (1, 2, 4, 8, 16, 24, 36, and 48 mg pr subject). For each patient, plasma samples were collected from before hour 0 to hour 48 post FG001 injection. All samples were analyzed (Labcorp Laboratories Limited) using an analytical method based on liquid chromatography with tandem mass spectrometric detection validated according to glucagon-like peptide. FG001, ICG-Glu-Glu-AE105.

FIGURE 7.

Comparison of 5-ALA and FG001 imaging of a high-grade glioma. Note the left part of the tumor that is located deep and accordingly is only shown on the FG001 image as NIR has a depth visualization of approximately 1 cm compared with <1 mm for blue light. The fluorescent signal in the vessels is a product of a long half-life of FG001. Image was obtained with Zeiss Pentero. FG001, ICG-Glu-Glu-AE105; NIR, near-infrared.