First-in-human intraoperative near-infrared fluorescence imaging of glioblastoma using cetuximab-IRDye800

Abstract

Introduction: Maximizing extent of surgical resection with the least morbidity remains critical for survival in glioblastoma patients, and we hypothesize that it can be improved by enhancements in intraoperative tumor detection. In a clinical study, we determined if therapeutic antibodies could be repurposed for intraoperative imaging during resection.

Methods: Fluorescently labeled cetuximab-IRDye800 was systemically administered to three patients 2 days prior to surgery. Near-infrared fluorescence imaging of tumor and histologically negative peri-tumoral tissue was performed intraoperatively and ex vivo. Fluorescence was measured as mean fluorescence intensity (MFI), and tumor-to-background ratios (TBRs) were calculated by comparing MFIs of tumor and histologically uninvolved tissue.

Results: The mean TBR was significantly higher in tumor tissue of contrast-enhancing (CE) tumors on preoperative imaging (4.0 ± 0.5) compared to non-CE tumors (1.2 ± 0.3; p = 0.02). The TBR was higher at a 100 mg dose than at 50 mg (4.3 vs. 3.6). The smallest detectable tumor volume in a closed-field setting was 70 mg with 50 mg of dye and 10 mg with 100 mg. On sections of paraffin embedded tissues, fluorescence positively correlated with histological evidence of tumor. Sensitivity and specificity of tumor fluorescence for viable tumor detection was calculated and fluorescence was found to be highly sensitive (73.0% for 50 mg dose, 98.2% for 100 mg dose) and specific (66.3% for 50 mg dose, 69.8% for 100 mg dose) for viable tumor tissue in CE tumors while normal peri-tumoral tissue showed minimal fluorescence.

Conclusion: This first-in-human study demonstrates the feasibility and safety of antibody based imaging for CE glioblastomas.

Keywords: Antibody-based imaging; Brain neoplasms; Cetuximab; Fluorescence; Glioblastoma; Image-guided surgery; Phase 1.

Fig. 1

Trial imaging workflow. Real-time fluorescent imaging was performed in the operating room (1) using a wide-field NIR imaging device on day 2 following cetuximab-IRDye800 infusion. 2 A closed-field NIR imaging system was used to image the resected specimens ex-vivo. 3 Specimens were sectioned and scanned in surgical pathology using a fluorescence scanning system to correlated fluorescence with histological tissue classification

Fig. 2

Representative intraoperative fluorescent images and associated pre-operative Magnetic Resonance Images (MRIs). Fluorescence image following tumor exposure in a patient 1, b patient 2, and c patient 3 and associated MRIs below

Fig. 3

Fluorescent images of serially cut fresh tumor and histologically normal tissue imaged in the closed-field setting. Serial sections from a patient 1 and b patient 2. Minimal detectable tumor weights were found to be 70 and 10 mg for patient 1 (low dose) and patient 2 (high dose), respectively

Fig. 4

MFI of tumor tissue compared to histologically normal and necrotic tumor tissue in glioblastoma patients. a MFI for patient 1 and patient 2 combined, b MFI of patient 1 only (low dose patient), c MFI of patient 2 only (high dose patient) (*P < 0.01, **P < 0.001), d TBR of patient 1 and 2 combined, patient 1 alone, patient 2 alone

Fig. 5

Representative images and statistical analysis of histological correlation with fluorescence. H&E stain, EGFR and fluorescent images of representative sections containing histologically normal tissue (a) and sections containing tumor and necrotic tissue (b). ROC curves for low (c) and high (d) dose CE glioblastoma patients and associated sensitivity and specificity calculations