Evaluation of nimotuzumab Fab2 as an optical imaging agent in EGFR positive cancers

Abstract

Molecular-targeted imaging probes can be used with a variety of imaging modalities to detect diseased tissues and guide their removal. EGFR is a useful biomarker for a variety of cancers, because it is expressed at high levels relative to normal tissues. Previously, we showed the anti-EGFR antibody nimotuzumab can be used as a positron emission tomography and fluorescent imaging probe for EGFR positive cancers in mice. These imaging probes are currently in clinical trials for PET imaging and image-guided surgery, respectively. One issue with using antibody probes for imaging is their long circulation time and slow tissue penetration, which requires patients to wait a few days after injection before imaging or surgery, multiple visits and longer radiation exposure. Here, we generated a Fab₂ fragment of nimotuzumab, by pepsin digestion and labeled it with IRDye800CW to evaluate its optical imaging properties. The Fab₂ had faster tumor accumulation and clearance in mice relative to the nimotuzumab IgG. The fluorescent signal peaked at 2 h post injection and remained high until 6 h post injection. The properties of the Fab₂ allow a higher signal to background to be obtained in a shorter time frame, reducing the wait time for imaging after probe infusion.

Figure 1

Nimotuzumab IgG pepsin digestion and Fab₂ labeling. Nimotuzumab IgG (150 kDa) was digested with pepsin resin. The Fab₂ (110 kDa) was purified with Pierce F(ab′)₂ Preparation Kit followed by size exclusion and labeled with IRDye800CW.

Figure 2

Analysis of Nimotuzumab pepsin digestion. (a) The percentage of IgG or fragments at different pepsin digestion time points are shown on the left y-axis and the sizes (kDa) are shown on the right y-axis. The column highlighted in grey shows the optimal time for pepsin digestion, which yields the most Fab₂. The pepsin digestion time in hours (h) is shown on the x-axis. (b) Electronic electrophoresis analysis of Fab₂ after pepsin digest and protein A clean up for IgG and Fc fragments. The peaks are labeled numerically, the first peak shows the 5 kDa lower marker, protein sizes are shown in kDa and the percent (%) of the total protein is shown. (c) Size exclusion analysis and purification of Fab₂ after pepsin digest of each fraction. The different colors represent the percentage of each fragments size from each fraction. (d) Electronic electrophoresis analysis of the final Fab₂ preparation. After pepsin digest, protein A clean up of IgG and Fc fragments, and size exclusion purification (collection of fractions 50–52).

Figure 3

Binding of nimotuzumab Fab₂ to cells and serum stability. Fab₂ and IRDye800CW-labeled Fab₂ were titrated with A-431 cells and analyzed by flow cytometry to obtain a binding constant to EGFR. (a) shows a representative histogram of the mean fluorescence intensity (MFI) of the unlabeled Fab₂ at each concentration tested. (b) shows the titration curves produced from plotting the concentration vs normalized MFI to obtain percent bound. (c) serum stability of IRDye800CW-labeled Fab₂ was tested at 37 °C, run on SDS-PAGE and analyzed for IRDye800CW fluorescence.

Figure 4

Optical imaging of IRDye800CW-labled nimotuzumab Fab₂ in A-431 xenografts. Mice bearing A-431 xenografts were injected with IRDye800CW-labeled nimotuzumab Fab₂ and imaged over time. (a) Dorsal and ventral mouse images taken over time. X = xenograft, B = bladder, L = liver, K = kidney. (b) Normalized signal of xenograft, kidney, liver and contralateral over time. (c) Tumor-to-background ratio (TBR) in the xenograft over time.

Figure 5

Optical imaging of IRDye800CW-labled nimotuzumab Fab₂ in control xenografts. Mice bearing H2009 and/or P3X63Ag8 xenografts were injected with IRDye800CW-labeled nimotuzumab Fab₂ and imaged over time. (a) Dorsal and ventral mouse images taken, 6 and 24 h images are shown. P = P2X63, H = H2009. (b) Normalized signal of control xenografts compared to A-431 EGFR positive xenografts are shown over time. ****p ≥ 0.0001, **p ≥ 0.01.

Figure 6

Comparison of nimotuzumab IgG and Fab₂ imaging in mice. (a) Normalized signal of IRDye800CW-labeled nimotuzumab IgG and Fab₂ imaging in the xenograft. The ratio of fluorescent signal of nimotuzumab Fab₂ to IgG over time in the xenograft (b), liver (c), and kidney (d) Errors are shown as standard deviation. Data from nimotuzumab IgG was previously published in and was re-analyzed and presented here as a comparison to the Fab₂. Significance shown on A-431 xenografts is compared to both controls. ****p ≥ 0.0001, **p ≥ 0.01.