Evaluation of a new optical imaging agent for c-Met in healthy volunteers and in subjects with high suspicion of colorectal cancer

L Schrier2, IMC de Visser-Kamerling2, AJ Kales2, M de Kam2, GT Dalsgaard1, P Gordon1, A Healey1, R Bendiksen1, G Torheim1, AMJ Langers3, J Morreau4, JCH Hardwick3, J Burggraaf2. 1GE Healthcare, N-0401, Norway, 2Centre for Human Drug Research, 2333 CL, The Netherlands, 3Leiden University Medical Center, Gastrointestinal disorders, 2333 ZA, The Netherlands, 4Leiden University Medical Center, Pathology, 2333 ZA, The Netherlands

The development of c-Met diagnostics to assess the expression of c-Met and to predict the response to treatment with targeted inhibitors is crucial for making good treatment decisions in colorectal cancer (CRC). There is good evidence that c-Met is over-expressed in pre-malignant and malignant CRC lesions. Early lesions over-expressing c-Met compared to normal surrounding colon may be detected by modified, red light-sensitive endoscopic equipment.

In this study, the safety and imaging properties of a new imaging agent which targets human c-Met was investigated in combination with standard white light (WL) and modified fluorescent light (FL) in 20 healthy volunteers (part 1) and in 15 subjects with high suspicion of colorectal cancer (part 2). Part 1 was a parallel-group, placebo-controlled, double-blind, randomized, single ascending-dose study to assess the safety of 4 dose levels (0.02, 0.04, 0.09 and 0.18 mg/kg bw) of the imaging agent. Twenty volunteers were enrolled and allocated to receive one of 4 ascending doses (or placebo; 0.9% NaCl) with 5 subjects in each of the 4 dose groups. Sigmoidoscopy was performed 4 times at hourly intervals post-injection (pi), and at each time point the operator made 8 “snap-shots” of the colon wall and 8 contact measurements. Fluorescence signal intensity was calculated for each region of interest to determine the appropriate dose and imaging time-point for the second part of the study, an open-label single-dose study in 15 subjects with high suspicion of CRC. In this part, subjects underwent colonoscopy after a single dose of the imaging agent and biopsies were taken from suspicious lesions. The imaging agent did not cause any clinically significant AEs or changes in symptoms, blood and urinary laboratory parameters, vital signs or ECG variables. Based on data from the first part, a dose of 0.13 mg/kg bw and imaging time point of 3h pi was selected for the second part, in which a total of 101 suspicious lesions were biopsied based on their appearance in WL and/or FL; 84 of these lesions were visible in WL and 89 had mildly or clearly increased fluorescence. Twelve of the 84 lesions seen in WL did not show increased fluorescence compared to background and 3 were not visible in FL. Seventeen (17%) of the 98 suspicious lesions visible with FL were not seen with WL and were characterized by small size and flat morphology; 82% of these 17 lesions were neoplastic in nature. The level of c-Met expression increased with the level of fluorescence and was higher in lesions exhibiting pre-malignant pathology than lesions showing normal histomorphology. Pharmacokinetic data showed that exposure (AUC) to the imaging agent in healthy volunteers was dose-linear, and that clearance was comparable between dose groups (~0.13 L/kg/h). In conclusion, use of the imaging agent in combination with WL and FL sigmoidoscopy/colonoscopy is feasible and well-tolerated in healthy volunteers and subjects with a high suspicion of CRC. The imaging agent shows dose-linear kinetics in the range of 0.02-0.18 mg/kg, and leads to fluorescence in polyps of all histological types that are thought to progress to malignancy. Use of this agent can enhance polyp detection during colonoscopy as it accumulates in c-Met-rich lesions. Further studies are needed to directly compare polyp detection by WL and FL colonoscopy using state-of-the-art, fluorescence-capable video endoscopy equipment, when available, rather than the optical fiber-based endoscopes used in this study, and randomization of the order of imaging mode during examination of the colon.