TUESDAY, Feb. 1 (HealthDay News) -- High levels of a certain protein in cancer cells may indicate which tumors are likely to spread, scientists report.
In a new study published in the current issue of the Journal of Clinical Investigation, high levels of a protein called CPE-delta N accurately predicted 90 percent of the time whether a tumor moved on from its original site in patients with liver cancer or some rare forms of adrenal cancer.
"We can tell from the levels [of the protein] whether the tumor has spread, and we can predict whether the tumor is likely to recur in the same tumor or tissue or in other parts of the body," senior study author Y. Peng Loh, of the U.S. National Institute of Child Health and Human Development, said during a Thursday news conference on the finding. "Currently, there are no accurate biomarkers that can achieve such predictions."
Cancer that has spread, or metastasized, is often fatal, and scientists have long tried to find ways to predict whether metastasis is likely to occur, not only to help guide and individualize treatment, but also to provide targets for potential future therapies.
Although it's still early, the results suggest it may be possible to develop tests that assess the likelihood of a cancer spreading and to treat it before it does so, the researchers say.
Doctors now use the stage and grade of a tumor to determine a patient's prognosis, and in this study those factors were far less accurate than the protein.
One expert found the finding intriguing.
"It's not entirely clear how it's working and how strong it's going to be in many different cancers, but it's the first time the gene has been linked to cancer," said Erica Golemis, deputy chief scientific officer at Fox Chase Cancer Center in Philadelphia. "This new spliced form has many cancer-promoting qualities and, at least in some liver cancers, it's doing something to support the aggressiveness of the disease."
Loh and colleagues from the United States, Hong Kong and Canada took tumor and tissue samples from 99 patients with liver cancer and 14 patients with pheochromocytoma, a tumor in the adrenal glands, and paraganglioma, another rare tumor found in the adrenal glands and sometimes elsewhere in the body.
When CPE-delta N levels reached a certain threshold -- twice that of surrounding tissue -- the cancer was very likely to spread or recur within two years. Below this threshold, chances that the cancer would reappear were much lower.
The protein was accurate in predicting which tumors would not recur within two years 76 percent of the time.
CPE-delta N is a variant of an enzyme, carboxypeptidase E (CPE), usually involved in processing hormones, Loh explained.
Besides beating out predictions based on stage and grade, CPE-delta N has the advantage of needing only "a small sample [of tissue or tumor] without full resection," said Dr. Stephen M. Hewitt, head of the Tissue Array Research Program of the U.S. National Cancer Institute.
Determining stage and grade usually requires much larger samples, often obtained through surgery.
CPE-delta N levels were also elevated in aggressive breast, colon and head and neck tumors, indicating that the measure might have utility across several cancers, something that would distinguish it from most other biomarkers identified so far.
But, Hewitt cautioned, "[CPE-delta N's] extension to other tumors would be speculative at best at this time. Biomarker development is very much like drug development. It takes time and sometimes even longer because we're waiting for a patient's outcome to determine if it works or not. It's not a short order of response in eight to 12 weeks."
Blocking the production of CPE-delta N levels in mice kept tumors from growing. But that research is extremely preliminary, and there is no way to use the same technique to block protein levels in humans at this point.
"We're studying what regulates expression of this gene and, if we can find what shuts it down, we can develop some small molecules that might shut down some other way," Loh said. "The bottom line is CPE-delta N is a potentially good target [for treatment]."
Loh said their group is now looking at the value of the protein as a prognostic marker in other types of cancer, including ovarian and perhaps kidney.
"This should be practice-changing after a bigger trial has been done and we're actively pursuing that [bigger trial]," she said.
The U.S. National Cancer Institute has more on metastatic cancer.
SOURCES: Jan. 27, 2011, teleconference with: Y. Peng Loh, Ph.D., senior investigator, Section on Cellular Neurobiology, U.S. National Institute of Child Health and Human Development, Bethesda, Md.; Stephen M. Hewitt, M.D., Ph.D., staff scientist, head, Tissue Array Research Program, U.S. National Cancer Institute, Bethesda, Md.; Erica Golemis, Ph.D., deputy chief scientific officer, Fox Chase Cancer Center, Philadelphia; Journal of Clinical Investigation
Copyright © 2011 HealthDay. All rights reserved.