IND-directed safety and biodistribution study of intravenously injected cetuximab-IRDye800 in cynomolgus macaques

Abstract

Purpose: The use of receptor-targeted antibodies conjugated to fluorophores is actively being explored for real-time imaging of disease states; however, the toxicity of the bioconjugate has not been assessed in non-human primates.

Procedures: To this end, the in vivo toxicity and pharmacokinetics of IRDye800 conjugated to cetuximab (cetuximab-IRDye800; 21 mg/kg; equivalent to 250 mg/m(2) human dose) were assessed in male cynomolgus monkeys over 15 days following intravenous injection and compared with an unlabeled cetuximab-dosed control group.

Results: Cetuximab-IRDye800 was well tolerated. There were no infusion reactions, adverse clinical signs, mortality, weight loss, or clinical histopathology findings. The plasma half-life for the cetuximab-IRDye800 and cetuximab groups was equivalent (2.5 days). The total recovered cetuximab-IRDye800 in all tissues at study termination was estimated to be 12 % of the total dose. Both cetuximab-IRDye800 and cetuximab groups showed increased QTc after dosing. The QTc for the cetuximab-dosed group returned to baseline by day 15, while the QTc of the cetuximab-IRDye800 remained elevated compared to baseline.

Conclusion: IRDye800 in low molar ratios does not significantly impact cetuximab half-life or result in organ toxicity. These studies support careful cardiac monitoring (ECG) for human studies using fluorescent dyes.

Figure 1

Cetuximab-IRDye800 imaging in macaques. White light (top) and fluorescent images (bottom) are shown of minor abrasions on the right eye of individual macaque in the cetuximab-IRDye800 group. The injuries to the eye were noted to have increased uptake of cetuximab-IRDye800, which returned to baseline by day 10 post infusion.

Figure 2

(a) Corrected QTc before and after treatment. ECG data (surface leads I, II, and II) was collected on cetuximab and cetuximab-IRDye800 groupsat 3 different days during acclimation (Base 1-3), pre-dose phase (Base 4), day 3 post infusion (Tx-Day 3), day 7 post infusion (Tx-Day 7), day 10 post infusion (Tx-Day 10), and day 15 post infusion (Tx-Day 15). QT interval data corrected using Bazett’s correction (called QTc = QT in msec/√R-R sec) was calculated on each electrocardiogram for all animals. (b) Corrected QTc during infusion. QT interval data corrected using Bazett’s correction (called QTc = QT in msec/√R-R sec) was calculated on each electrocardiogram for all animals at pre-dose (base) and every 10 minutes during the 1 hour infusion of cetuximab or cetuximab-IRDye800 groups. Values are mean QTc values ± standard error.

Figure 3

Biodistribution in tissues. Background fluorescence and cetuximab-IRDye800 fluorescence is reported for all organs collected during whole-body biodistribution at day 15 post infusion. Inset graph shows a standard curve from fluorescence levels versus known levels of cetuximab-IRDye800. Values are reported as percent-injected dose (%ID) per gram of tissue ± standard error.

Figure 4

Estimated total micrograms (μg) of cetuximab-IRDye800 in selected tissues. Background fluorescence and cetuximab-IRDye800 fluorescence is reported for selected tissues collected during whole-body biodistribution at day 15 post infusion. Values are reported as estimated total μg (mean) ± standard error.

Figure 5

Cetuximab and cetuximab-IRDye800 (mean μgmL) in the plasma over time as determined by Western blotting. (a) Panels show immunoblot of blood plasma from individual macaques (n=4) in the cetuximab-unlabelled group (U) and macaques (n=4) in the cetuximab-IRdye800 group (IR) imaged. A known amount of cetuximab-IRDye800 was also included (*). In the upper panel, bands represent presence of Goat anti-human IRDye 680RD detected at the 150kDa marker during Pearl Impulse acquisition of the membrane at the 700 nm channel. Lower panel shows fused image of Pearl Impulse acquisition using the 700 nm and 800 nm acquisition channel to verify cetuximab-IRDye800 at the 150kDa marker. (b)A plot of the natural log of the mean blood level versus time yielded a line with a slope equal to −k, and with the half-life equal to the Ln 2 divided by k. Values are mean μgmL ± standard error.