Researchers have created a new imaging compound in mice that selectively binds to certain cancer cells and glows, or fluoresces, only when processed by these cells. This cancer-specific fluorescence allowed the investigators to successfully visualize very small tumors in the peritoneum -- the tissue that lines the wall of the abdomen -- in mice with ovarian cancer. The sensitivity -- or ability to accurately detect small clusters of tumor cells -- of this approach was 92 percent. The study, conducted by researchers at the National Cancer Institute (NCI), part of the National Institutes of Health (NIH), and colleagues, appears in the March 15, 2007 issue of Cancer Research. A second study published in the April 15, 2007 issue of Cancer Research, showed even greater sensitivity with a compound designed to fluoresce while bound to the surface of cancer cells. In a third study, published in the November 1, 2007 issue of Clinical Cancer Research, researchers developed and tested a new imaging compound with similar properties that could be safer for use in humans.

"The virtue of these studies is that other fluorescent compounds have been tested for the detection of small clusters of cancer cells that might otherwise be missed during surgery, but those have drawbacks, such as being always fluorescent, which makes it difficult to distinguish tumor cells from normal tissue. These studies point to a potential solution to this problem," said NIH Director Elias A. Zerhouni, M.D.

"A fluorescent imaging compound that is specific for cancer cells holds great promise for the treatment of cancers, such as ovarian and pancreatic cancer, which often metastasize widely before diagnosis. In the coming years, as cancer research is increasingly based on an understanding of tumors down to a detailed molecular level, advanced imaging will be a key component of essentially every study," said NCI Director John E. Niederhuber, M.D.

The researchers, led by Hisataka Kobayashi, M.D., Ph.D., from NCI's Molecular Imaging Program in the Center for Cancer Research (CCR), first created a compound to be tested only in mice that consisted of the protein avidin, which binds to another protein commonly found on the surface of cancer cells that potentially can spread, or metastasize, to the peritoneum. They joined this compound to three molecules of the fluorescent compound rhodamine X. In this new compound, which they called Av-3ROX, the rhodamine X molecules are unable to fluoresce. However, when Av-3ROX is taken up by cancer cells after binding to them, it is broken down in sac-like compartments inside the cells called lysosomes. When enzymes in the lysosomes break the compound into smaller pieces, the rhodamine X is released and is able to fluoresce.

"Conventional imaging methods such as nuclear isotopes, MRI, or CT use contrast agents that make a signal whether they are bound or unbound to a cancer cell," said Kobayashi. "Using Av-3ROX, we only make a signal from cancer cells. It's cancer-specific imaging."

When the researchers injected the 'always on' fluorescent molecule Av-0.5ROX into the peritoneum of tumor-bearing mice, fluorescence was immediately detectable and more intense than that produced by Av-3ROX. However, the fluorescence produced by Av-0.5ROX lacked specificity, producing fluorescence in both tumor cells and the surrounding tissue, making it difficult to distinguish the tumor cells. In contrast, by three hours after Av-3ROX injection, the fluorescence intensity in normal tissues was less than with Av-0.5ROX, but the fluorescence intensity in tumor nodules was much higher than with Av-0.5ROX.

To confirm that Av-3ROX was primarily processed by tumor cells, the researchers performed a second experiment in mice, this time using cells that carried the gene for red fluorescent protein (RFP) to induce the initial tumors and peritoneal metastases. This approach allowed every metastasis to be detected using a camera and filter specific for RFP. The investigators then injected Av-3ROX into the peritoneum of the mice and captured fluorescent images of both Av-3ROX and RFP. Next, they compared the number of metastases identified using both compounds.

Out of 507 metastases, at least 0.8 millimeters in diameter, shown by RFP, Av-3ROX detected 465 of them, indicating a sensitivity of 92 percent. Only 2 percent of metastases identified by Av-3ROX turned out to be false positives, translating to a 98-percent tumor detection accuracy, or specificity, for this technique.

Although the data provide proof-of-concept for this type of molecular imaging technique, Av-3ROX cannot be used in people, because the avidin portion of the compound would cause an immune system reaction. Kobayashi and his colleagues worked on a second-generation compound that joins the binding site of avidin -- the part that recognizes the cancer cells -- to human serum albumin. This compound "should not create a harmful immune response because it's based on a human protein," said Kobayashi.

In the second study, the researchers designed a cancer cell-specific imaging compound that fluoresces while bound to a receptor on the outside of cancer cells. The imaging compound consists of two parts: a monoclonal antibody attached to more than 10 copies of a molecule called biotin, and a fluorescent molecule nAv-BDPfl, which glows approximately 10 times brighter when bound to biotin. A monoclonal antibody is a type of protein made in the laboratory that can locate and bind to substances on the surface of cancer cells.

The researchers designed a procedure to test the ability of this imaging compound to pinpoint small metastases in the peritoneum of mice with ovarian cancer. They found that the two-step imaging method had a sensitivity of 96 percent for metastases at least 0.8 millimeters in diameter, and a 98 percent tumor specificity. It is not yet known if this compound can be used "as is" in people.

Next the Kobayashi team embarked on a third study to test a new compound (GmSA-20ROX). A version of this new compound that had been radio-labeled (joined with a radioactive atom) has already been tested in humans and targets the same ovarian cancer receptors as did the team's earlier imaging compounds. Again using mouse models, the researchers found that this new compound binds to the cancer cells more efficiently and quickly than avidin. Within one hour after the compound was introduced, both the sensitivity and the specificity of this imaging approach were 99 percent. The researchers believe that, in the future, GmSA-20ROX could be used, either alone or in combination with special cameras and filters, to assist surgeons trying to remove ovarian cancer tumors that have spread beyond the original site.

Depending on the type of cellular receptor targeted in different cancers, a molecular imaging compound, once it attaches to its receptor, may be taken into the cell or remain bound to the surface. Kobayashi and his colleagues continue to be interested in developing different compounds that can fluoresce in either location. They believe that their approach to developing new imaging agents holds promise in optically enhancing surgical or endoscopic procedures, allowing for more complete surgical removal of metastatic disease.

###

Hama Y, Urano Y, Koyama Y, Kamiya M, Bernardo M, Paik RC, Shin IS, Paik CH, Choyke PL, Kobayashi H. A target cell-specific activatable fluorescence probe for in vivo molecular imaging of cancer based on a self-quenched avidin-rhodamine conjugate. Cancer Research. Vol. 67, No. 6, March 15, 2007.

Hama Y, Urano Y, Koyama Y, Choyke P, Kobayashi H. Activatable fluorescent molecular imaging of peritoneal metastases following pretargeting with a biotinylated monoclonal antibody. Cancer Research, Vol. 67, No. 8, April 15, 2007.

Hama Y, Urano Y, Koyama Y, Gunn A, Choyke P, Kobayashi H. A Self-Quenched Galactosamine-Serum Albumin-RhodamineX Conjugate: A "Smart" Fluorescent Molecular Imaging Probe Synthesized with Clinically Applicable Material for Detecting Peritoneal Ovarian Cancer Metastases. Clinical Cancer Research, Vol. 13, No. 21, November 1, 2007.

Researchers are from the Molecular Imaging Program, Center for Cancer Research, NCI, Bethesda, Md.; the Nuclear Medicine Department, Warren Magnuson Clinical Center, NIH, Bethesda, Md.; and the Graduate School of Pharmaceutical Sciences, the University of Tokyo, Japan.

For more information on NCI's Molecular Imaging Program, including Drs. Kobayashi, Choyke, and Bernardo, go to http://ccr.cancer.gov/labs/lab.asp?labid=175.

For more information about cancer, please visit the NCI Web site at http://www.cancer.gov, or call NCI's Cancer Information Service at 1-800-4-CANCER (1-800-422-6237).

Back to Top