COX1 and COX2 May Both Contribute to Intestinal Tumors

NEW ORLEANS--Colon cancer-prone mice bred to lack the cyclooxy-genase 1 (COX1) enzyme have the same reduction in tumors as those bred to lack the cyclooxygenase 2 (COX2) enzyme. This result suggests that both forms of COX may encourage tumor development, according to research reported by Patricia Chulada, PhD, at the 89th annual meeting of the American Association for Cancer Research (AACR).

Drugs that inhibit COX are under investigation as anticancer agents because of some intriguing results. People who routinely take aspirin may reduce their risk of colon cancer by as much as half, Dr. Chulada, a toxicologist at the National Institute of Environmental Health Sciences, said at a press briefing. Also, another nonsteroidal anti-inflammatory drug (NSAID), sulindac, has been shown to reverse polyps in people with familial adenomatous polyposis (FAP), an inherited disorder in which individuals develop hundreds of intestinal polyps.

One theory is that aspirin, sulindac, and perhaps other NSAIDs interfere with cancer development by inhibiting COX, Dr. Chulada said. This enzyme is involved in the production of prostaglandins, which are necessary for inflammation and play a role in cancer.

Side Effects Limit NSAID Use

But side effects limit the routine use of NSAIDs. Inhibiting the near-ubiquitous COX1 enzyme can damage the stomach lining and kidney. In contrast, COX2 is produced mainly in pathologic conditions such as cancer, and its inhibition causes fewer side effects.

Researchers are working on agents that inhibit COX2 but not COX1 in the hope that such drugs would retain their potency against cancer and arthritis, yet not cause ulcers or kidney disease.

But the work of Dr. Chulada and her co-workers suggests that COX1 may also contribute to the development of tumors. She noted, for example, that COX1 can enhance blood vessel formation in tumors and can promote metastasis.

In her study, Dr. Chulada interbred three types of mice:

• Min C57BL-6J mice, a strain with an FAP-like condition in which each mouse naturally develops 45 to 57 tumors in the small intestine and a few tumors in the colon.
• COX1-null mice, a strain that lacks the COX1 gene and so does not produce COX1.
• COX2-null mice, a strain that lacks the COX2 gene and so does not produce COX2.

The resulting mice were genetically prone to form tumors but lacked one or the other form of COX.

Both May Play a Role

Dr. Chulada and her colleagues found that whether the experimental mice lacked COX1 or COX2, they had only one-fifth to one-tenth as many tumors as is usual in a Min mouse. This result suggests that both COX1 and COX2 play a role in the development of intestinal tumors in this mouse model.

The message, Dr. Chulada said later in an interview with Oncology News International, is that researchers "need to look at the mechanisms of exactly how COX1 and COX2 are involved in carcinogene-sis." Aspirin and sulindac may be so effective against polyps because they inhibit both forms of COX rather than just one, she said.

"It is too soon to rule out a role for COX1 in cancer development," Dr. Chulada noted, "and it will be important to test new COX2 inhibitors to see how effective they are in reducing the risk of colon cancer."