In Vivo Fluorescence Immunohistochemistry: Localization of Fluorescently Labeled Cetuximab in Squamous Cell Carcinomas

Abstract

Anti-EGFR (epidermal growth factor receptor) antibody based treatment strategies have been successfully implemented in head and neck squamous cell carcinoma (HNSCC). Unfortunately, predicting an accurate and reliable therapeutic response remains a challenge on a per-patient basis. Although significant efforts have been invested in understanding EGFR-mediated changes in cell signaling related to treatment efficacy, the delivery and histological localization in (peri-)tumoral compartments of antibody-based therapeutics in human tumors is poorly understood nor ever made visible. In this first in-human study of a systemically administered near-infrared (NIR) fluorescently labeled therapeutic antibody, cetuximab-IRDye800CW (2.5 mg/m(2), 25 mg/m(2), and 62.5 mg/m(2)), we show that by optical molecular imaging (i.e. denominated as In vivo Fluorescence Immunohistochemistry) we were able to evaluate localization of fluorescently labeled cetuximab. Clearly, optical molecular imaging with fluorescently labeled antibodies correlating morphological (peri-)tumoral characteristics to levels of antibody delivery, may improve treatment paradigms based on understanding true tumoral antibody delivery.

Figure 1

Localization of cetuximab-IRDye800CW in histologic sections of tissue with tonsil, tongue and cutaneous squamous cell carcinoma (SCC). (a) Representative 150 kD (=intact cetuximab-IRDye800CW) and 1kD (=IRDye800CW) bands are shown by SDS-PAGE of individual patient blood samples of each dosage cohort. (b) Representative haemotoxylin/eosin (H&E), cytokeratin and fluorescence image of SCC sections of tonsil, tongue and cutaneous origin. (c) Mean Fluorescence Intensity (MFI) quantified in pathology-positive areas of tumor and healthy surrounding for nine patients (n = 3 at each dosage cohort) for respectively 2.5 mg/m², 25 mg/m², and 62.5 mg/m². Data are presented as mean ± SD, ^* = P < 0,001. Scale bars in all images represent 100 μm.

Figure 2

Co-localization of fluorescence signals of cetuximab-IRDye800CW and Epidermal Growth Factor Receptor (EGFR) expression. (a) Representative haemotoxylin/eosin (H&E) image indicating tumor (T) and normal (N) with corresponding EGFR expression immunohistochemistry stain and fluorescence image. (b) Increase in Mean Fluorescence Intensity (MFI) as a function of EGFR Density (% area EGFR positive). Data are presented as mean ± SD, * = P < 0,001. Scale bars in all images represent 100 μm.

Figure 3

Fluorescence from adjacent normal tissue correlates with distance from tumor.Representative H&E image tumor (T) and normal (N) regions outlined by a board-certified pathologist (a). Signal-to-tumor ratio is determined at predefined distances from tumor (1-5 mm) as depicted by rectangular regions-of-interest (ROI) at 0, 1, 2, 3, 4 and 5 mm margin distance (b) graphed (c). Data are presented as mean ± SD. Scale bar represent 50 μm.

Figure 4. Effect of tissue properties on cetuximab-IRDye800CW uptake.

(a) Representative haemotoxylin/eosin (H&E) image and corresponding fluorescence image. Differentiated keratinizing cancer cells are outlined by black arrows. (b) White arrows outline tumor necrosis. (c) Regression analysis between % Area keratinized and Mean Fluorescence Intensity (MFI) (P < 0.003). (d) Regression analysis between % Area necrosis and Mean Fluorescence Intensity (MFI). Data are presented as mean ± SD, ^* = P < 0,001. Scale bars in all images represent 100 μm.