NEW YORK CITY—Cyclooxygenase (COX)-2 inhibitors, traditionally used to treat arthritis pain and inflammation and now showing promise in the prevention and treatment of cancer, especially those of the colon and lung, are a good example of molecular targeted therapy. So said Andrew J. Dannenberg, MD, Director of Cancer Prevention at New York Weill Cornell Medical Center, speaking at a news briefing here called “From Arthritis to Cancer—A New View of COX-2.”
The COX-2 system appears to play a role in a number of pathways to cancer, another expert, David Weinberg, MD, Director of Gastroenterology at Fox Chase Cancer Center in Philadelphia, noted in an interview.
Inhibition of the COX-1 system, on the other hand, is responsible for some of the unwanted side effects, specifically gastrointestinal bleeding, from non-selective non-steroidal anti-inflammatory drugs (NSAIDS), he explained. “Hence, COX-2 is targeted in the sense that it preferentially affects the bad pathways while sparing the good.”
COX-2 cannot be detected in most normal tissues, but is induced by oncogenes and overexpressed in premalignant and malignant tumors, Dr. Dannenberg said. Consequently, COX-2 inhibitors may help to target the earliest stages of carcinogenesis, he said.
In colorectal cancer, COX-2 is overexpressed in 85 percent of adenomas. The enzyme is also overexpressed in gastric cancer, Barrett's esophagus, hepatocellular cancer, pancreatic malignancies, oral leukoplakia, head and neck cancer, non-small cell lung cancer, and breast, prostate, and bladder cancers.
Specifically, overexpression of COX-2 inhibits apoptosis, allowing cancer to grow, and COX-2 inhibitors can help ensure that cancer cells die. For example, a 2000 study found that the COX-2 inhibitor celecoxib induces apoptosis by blocking the activation of the key antiapoptotic kinase Akt in human prostate cancer cells independently of the antiapoptotic gene Bcl-2 (Hsu AL et al: J Biol Chem 2000;275: 11397-11403).
Exposure of these cells to celecoxib induced such features of apoptosis as morphologic changes, DNA “laddering” (a characteristic pattern of discontinuous DNA fragments), and activation of the key apoptotic mediator caspase-3. However, piroxicam, a COX-1-specific inhibitor, had no noticeable effect on cancer cells after prolonged exposure.
Also speaking at the news briefing, William Li, MD, President and Medical Director of the Angiogenesis Foundation in Cambridge, MA, noted that overexpression of COX-2 also contributes to inflammation and angiogenesis, processes inextricably involved in cancer. COX-2 inhibitors are both anti-inflammatory and antiangiogenic.
The antiangiogenic properties of COX-2 inhibitors may be independent of COX-2 expression in tumors, Dr. Li added. “Because virtually all tumors undergo infiltration by inflammatory cells such as monocytes and macrophages, COX-2 inhibitors could help to suppress the angiogenesis promulgated by these pre-tumoral cell elements.”
Traditional NSAIDs such as ibuprofen inhibit both COX-1 and COX-2. COX-1 is expressed in most tissues, healthy or otherwise, and converts to prostaglandins that help protect the gastric mucosa. Consequently, inhibiting COX-1 in addition to COX-2 can cause adverse side effects such as gastrointestinal bleeding.
Celecoxib vs Naproxen
Dr. Dannenberg pointed to a 1999 study (JAMA 1999;282:1921–1928) where researchers at Beth Israel Deaconess Medical Center of Harvard Medical School tested whether celecoxib has efficacy as an anti-inflammatory and analgesic agent with reduced gastrointestinal tract mucosal damage compared with conventional NSAIDs in patients with rheumatoid arthritis.
This randomized, multicenter, placebo-controlled, double-blind trial included 1,149 patients aged 18 years or older with symptomatic rheumatoid arthritis. Patients received celecoxib at 100, 200, or 400 mg twice daily or naproxen at 500 mg twice daily.
The incidence of endoscopically determined gastroduodenal ulcers in placebo-treated patients was four percent, and the incidences across all dosages of celecoxib were not significantly different. In contrast, the incidence with naproxen was 26 percent.
Celecoxib vs Ibuprofen & Diclofenac
Additionally the CLASS (Celecoxib Long-term Arthritis Safety Study) by Silverstein et al (JAMA 2000;284:1247–1255) found that arthritis patients taking this COX-2 specific inhibitor at four times the recommended osteoarthritis dose experienced fewer symptomatic gastrointestinal ulcers and ulcer complications than did patients taking ibuprofen and diclofenac. Furthermore, celecoxib showed no increase in thromboembolic or other cardiovascular-related events, even among non-aspirin users.
In comparison to Celebrex, ibuprofen and diclofenac were associated with a significantly greater gastrointestinal blood loss—equivalent to two pints or more—over the course of the study, even in patients not experiencing bleeding ulcers.
The study was an approximately 13-month, multicenter, randomized, double-blind outcomes trial of about 8,000 arthritis patients—5,800 with osteoarthritis and 2,200 with rheumatoid arthritis. Some patients were taking low-dose aspirin for cardioprotection.
Low Side Effects Lead to Interest in Selective Inhibitors
The therapeutic benefit of NSAIDs is due to the inhibition of COX-2, Dr. Dannenberg said, adding that the toxicity of these drugs is due to inhibition of COX-1. For this reason, NSAIDs have not been commonly used in humans to prevent colon cancer.
“A low to moderate risk is needed to be considered for prevention,” he said. However, selective NSAIDs.i.e., those that block only COX-2—including rofecoxib and celecoxib, are of great interest to researchers partly because they don't carry significant side effects, said Mark Pochapin, MD, Director of Gastrointestinal Endoscopy at Weill Medical College.
Modest differences exist between celecoxib and rofecoxib, Dr. Weinberg noted. “How much difference that makes [in prevention and treatment] is unknown,” he said.
Rofecoxib and celecoxib are both effective in mouse models, said Ernest T. Hawk, MD, Chief of the Gastrointestinal & Other Cancers Research Group in the National Cancer Institute's Division of Cancer Prevention. Another COX-2 inhibitor called NS398 is also effective in mice, and Nimesulide, which is being studied in Europe, is effective in animal models, he noted.
Celecoxib for Cancer Prevention
“Celecoxib looks promising as a preventive agent,” he said, adding that the drug has clear effects on colorectal adenomas.
For example, a 2000 trial sponsored by the NCI Division of Cancer Prevention in collaboration with Searle, was the first study to explore the possible side effects of celecoxib in reducing polyps among people with familial adenomatous polyps (Steinbach et al: N Engl J Med 2000;342:1946–1952).
This randomized, double-blind, placebo-controlled trial studied 83 patients with familial adenomatous polyposis (FAP) and found that 400 mg of celecoxib twice daily for six months significantly reduced the number of adenomatous colorectal polyps by an average of 28 percent, compared with only five percent reduction with placebo.
In December 1999, the Food and Drug Administration approved celecoxib as an adjunctive therapy for FAP. That autosomal dominant disorder provides researchers with insight into the biology of adenomas and colorectal cancer development, Dr. Hawk noted.
Some 15 percent of patients with FAP develop colon cancer by age 10, and by the age of 40, nearly all have cancer. The disease accounts for only one percent of colorectal cancers in the United States.
Treating Sporadic Adenomatous Polyps (SAP)
Researchers are also interested in studying sporadic adenomatous polyps (SAP), which develop in patients without a genetic predisposition for polyp growth. SAP develops after age 50 in most patients, and 65 to 85 percent of colon and rectal cancers can be attributed to SAP growth.
Adenomas typically take 10 to 15 years to become cancerous, Dr. Pochapin explained. While not all polyps turn cancerous, the thinking is that all colon cancer develop from these growths—If a patient does not have polyps, colon cancer will not develop, he said. This is why using medication to prevent the formation of polyps is so important.
Dr. Pochapin and his colleagues are studying celecoxib and its effect on preventing sporadic adenomas in a Phase III, randomized, prospective, double-blind multicenter study comparing three regimens: celecoxib at 200 mg given twice daily, celecoxib at 400 mg given twice daily, or placebo. Patients must have one large polyp or multiple polyps of any size to participate. Limited aspirin use for the heart is allowed.
Researchers are still determining how effective COX-2 inhibitors may be in preventing the development of cancer, Dr. Dannenberg said, and the hope is that the drug may be applicable to cancers other than those of the colon.
HER-2/neu Breast Cancer
Celecoxib may also prove to play a role in treating patients with HER-2/neu-positive breast cancer. Dr. Dannenberg and his colleagues studied 29 human breast cancers, 15 of which were HER-2/neu positive. Of these, 14 were also COX-2 positive. Only four of 14 HER-2/neu-negative cancers were COX-2 positive, and even these were not nearly as positive as those in the HER-2/neu-positive group, Dr. Dannenberg noted.
“There's a clear correlation between HER-2 status and COX-2 positivity,” he said, adding that COX-2 may be a molecular target in treating HER-2/neu-positive breast cancer.
In animal models, overexpression of COX-2 is sufficient to cause breast cancer. For example, he explained, in a 2000 article published in Cancer Research (Harris RE et al: 2000;60:2101–2103), researchers from Ohio State University College of Medicine studied celecoxib against mammary cancer in female rats. Treatment with celecoxib was compared with treatment with ibuprofen and with a control group receiving dimethylbenz(a)anthracene.
Celecoxib (1,500 ppm) produced reductions in the incidence, multiplicity, and volume of breast tumors relative to the control group—68, 86, and 81 percent, respectively. Ibuprofen also produced significant effects, but of lesser magnitude—40, 52, and 57 percent, respectively.
Nasser Altorki, MD, Director of the Division of Thoracic Surgery at Weill, is studying COX-2 inhibitors for the treatment of lung cancer, in which COX-2 is often overexpressed.
In animal models, Dr. Altorki noted, celecoxib retards the growth of lung tumors and decreases the number and size of metastases. Furthermore, adding COX-2 to chemotherapy provides an additive effect in animal studies.
His group is conducting the first trial of COX-2 inhibitors to treat lung cancer and will present their data next month at this year's American Society of Clinical Oncology Annual Meeting. The results look promising, Dr. Altorki noted, citing the case of one patient who had a 90% to 95% reduction in a 5-cm lung tumor.
Treatment for NSCLC in the last 40 years has been surgery, chemotherapy, and radiation, but survival rates over this time have not greatly improved, he said. “We need to stop using shot guns to kill flies.”
The Angiogenesis Foundation is conducting studies at Brigham & Women's Hospital in Boston and has found that celecoxib can suppress the recurrence of skin cancer in patients with a history of multiple non-melanoma malignancies, Dr. Li reported.
“The recurrence rate appears to be suppressed dramatically, from one new skin cancer every two months to one every 12 to 18 months.”
The NCI is also investigating celecoxib in Phase I and II clinical trials alone and in combination with other therapies as a potential treatment for not only breast and lung cancers, but also prostate and cervical cancers.
Specifically, researchers are conducting a Phase I/II study of external-beam radiotherapy and brachytherapy concurrently with celecoxib, fluorouracil, and cisplatin in 62 women with locally advanced cervical cancer. Investigators want to determine whether this combination regimen increases locoregional control rates, distant control, disease-free survival, and overall survival in these patients
A randomized Phase I trial of neoadjuvant celecoxib followed by prostatectomy is underway in 60 to 70 men with localized prostate cancer. Investigators want to compare prostaglandin levels in tissue samples of those treated with neoadjuvant celecoxib versus a placebo followed by prostatectomy.
Researchers are also comparing the effect of these regimens on angiogenic factors within the prostate and determining the pharmacokinetic and pharmacodynamic effects of celecoxib.
Looking to the Future
Combination therapy for the treatment and prevention of cancer needs to be investigated further, Dr. Dannenberg remarked. Low doses of combinations of drugs may be more effective than when used alone and carry fewer side effects. Combinations might include COX-2 inhibitors with epidermal growth factor receptor kinase inhibitors or with retinoids.
The mechanisms by which COX-2 deters apoptosis and spurs angiogenesis and invasiveness need to be studied further because they may vary from tumor type to tumor type, he added.
Because COX-2 inhibitors appear to reduce the growth rate of tumors rather than causing them to regress in animal models, the drugs may be most beneficial in combination with chemotherapy or radiotherapy, he concluded.