br Optimal schedule of administration of Claudin
Optimal schedule of administration of Claudin-1-IRDye800CW
An initial time response study demonstrated no enhancement of tumor visualization compared with the background signal when mice were imaged 24 h after administration of the antibody-fluorophore conjugate. At 48 h, tumor margins were clearly defined with no surrounding background signal. 72 h after administrations of Claudin-1-IRDye800CW, there was no longer a tumor signal, suggesting metabolism of the dye at this time point (Fig. 2). Increasing amounts of Claudin-1-IRDye800CW were administered to mice (12.5, 25 and 50 mg),
Fig. 1 e Western blot of multiple cell-line and patient-derived primary and metastatic colon cancer cell lysates. The bottom line represents the control with B actin. LS174T is a colon cancer cell line. CM 1, 2, 3, and 6, liver 2, liver 6, lung 3, and PM9 are patient-derived colon cancer metastases. C4 is a patient-derived colon cancer primary tumor. Eight of nine tumor lysates demonstrated overexpression of Claudin-1 to varying degrees.
Fig. 2 e Nine mice were imaged at different time points to determine optimal time of imaging after administration of Claudin-1-IRDye800CW (A) 24 h, (B) 48 h, and (C) 72 h. Each mouse in the panel was administered 25 mg Claudin-1-IRDye800CW, and different mice were imaged at the various time points. The best tumor visibility with minimal background was 48 h after injection of Claudin-1-IRDye800CW. Arrows point to tumors on the bilateral flanks. (Color version of figure is available online.)
and demonstrated clear tumor margins at each dose when imaged 48 h after administration. For subsequent studies, mice received 25 mg Claudin-1-IRDye800CW via tail vein in-jection and mice were imaged 48 h after administration.
In vivo fluorescence labeling of Claudin-1 ABT263 in orthotopic and carcinomatosis models
All orthotopic mouse models, including cell line and PDOXs, demonstrated fluorescence of tumors with clear margins after administration of the fluorescent anti-Claudin antibody (Fig. 3). There was a minimal background signal as seen on fluorescence images. All models demonstrated fluorescence of the liver due to metabolism of the dye in the liver (Fig. 3). Three weeks after intraperitoneal injection of LS174T human colon cancer cells, mice were imaged to assess for peritoneal im-plantation and tumor burden. Each mouse demonstrated multiple small fluorescent tumors on the peritoneal surface of the abdominal wall and various organs (Fig. 4). Regional fluo-rescent metastases to the cecum and colon were identified in multiple PDOX models. Claudin-1-IRDye800CW was able to detect these small local metastases that were not visible under bright light (Fig. 5). The mouse that received Claudin-1 anti-body alone without conjugation to dye did not demonstrate any fluorescent signal on imaging with the Pearl Trilogy. The mouse that received IR-800CW dye alone without antibody did
not demonstrate tumor specificity and had a significant back-ground signal throughout the mouse.
Toxicity and side effects
After imaging was performed, mice were euthanized using techniques described previously and necropsy was per-formed. Internal organs were removed and examined to determine if any gross toxicity was identified. In all mice involved in the study, there were no gross defects of internal organs to suggest toxicity. All mice survived tail vein injection and administration of all doses. No mice had to be euthanized because of toxicity or side effects.
The results of the present study show that Claudin-1 is a useful target for NIR antibody-based imaging for visualization of colorectal tumors. Western blotting demonstrated over-expression of claudin to varying degrees in eight of the nine cell line and patient-derived colon cancer samples studied. Although not all samples overexpressed Claudin-1, it can be extrapolated that most colorectal tumors overexpress Claudin-1. The cell lysates with the highest signal intensity on Western blot were all derived from patient samples of colon cancer metastases. This is consistent with previous studies,
Fig. 3 e (A) Orthotopic model of colon cancer cell line LS174T. (B) PDOX model of C4, a patient-derived primary colon cancer. The tumor is brightly labeled with the fluorophore-antibody conjugate and the tumor margins are clearly defined in both mouse models, suggesting potential for use in surgical practice. Liver fluorescence is noted in both mice because of metabolism of the IR800 dye. The bladder is fluorescent due to excretion of the dye in the urine in panel (B). (Color version of figure is available online.)