The COX-2 Story: Is Any Drug Completely “Safe?”

Abstract

About 12 years ago, my laboratory published an article in Gastroenterology documenting increased expression of COX-2 in colorectal cancers, without changes in COX-1 levels when comparing normal matched colonic tissue with colorectal cancers or adenomas (Gastroenterology 1994;107:1183–1188). Our initial observation was followed by several reports from other laboratories that confirmed increased COX-2 expression in colorectal cancers. These results were preceded by many observational studies indicating a decreased risk of colorectal cancer in people who consumed NSAIDs on a regular basis (Nat Rev Cancer 2001;1:11–21), although even today the precise dose and duration of NSAID use required for optimal risk reduction is debated. Most NSAIDs inhibit both the COX-1 and COX-2 enzymes, with COX-2 thought to be the primary anti-inflammatory target. The connection between COX-2 and colorectal cancer led many scientists and clinicians to investigate whether inhibition of this pathway could provide some reduction in risk for colorectal and other cancers. In the early 1990s, the pharmaceutical industry began developing agents that specifically inhibited the COX-2 enzyme without affecting COX-1 activity. Although these COX-2 inhibitors were initially developed and approved for use in treating patients with arthritis or pain, their appearance opened up the possibility of evaluating these drugs for a role in cancer prevention as well.The first clinical trial on cancer evaluated the use of COX-2 inhibitors in patients with familial adenomatous polyposis (FAP). The clinical endpoint measured was polyp size and number. After 6 months of treatment, a significant reduction in polyp burden was observed in those FAP patients taking the highest dose of drug (N Engl J Med 2000;342:1946–1952). These results led to the FDA approval for use of Celecoxib (Celebrex) in patients with FAP. This study also led to the design of large trials studying subjects who develop sporadic polyps to determine whether COX-2 inhibitors could also prevent polyp recurrence in that group. At least 3 trials were designed and initiated. One trial evaluated the efficacy of Rofecoxib (Vioxx). The other 2 trials evaluated the efficacy of Celecoxib. The design of these trials required continuous treatment for 3 years, with a 2-year extension to evaluate drug safety. Investigators evaluated subjects who had previously undergone a complete colonoscopy with polypectomy and were randomized to treatment or placebo arms. Drug safety was carefully monitored because these trials were designed to observe a large contingent of study patients over a long period of time. Most of the arthritis and GI safety trials conducted previously were no longer than 12 months in duration. The results of the “cancer prevention” trials received an enormous amount of worldwide media attention mainly because of safety concerns. The first trial results to emerge involved the Rofecoxib study, which demonstrated an increased risk of thrombotic events after several months of treatment in patients on the drug (N Engl J Med 2005;352:1092–1102). This drug did show efficacy with regard to causing a reduction in polyp recurrence. However, the safety concerns led to the removal of Rofecoxib from the market and a landslide of litigation brought forward by people who were exposed to the drug and experienced a thrombotic or cardiovascular event. These side effects would not likely have been detected in a 12-month study, since there was no significant difference between the placebo and treated groups during that time frame. Next, the safety results from one of the Celecoxib trials emerged, which again demonstrated cardiovascular side effects after several months of treatment, especially in the high-dose group (N Engl J Med 2005;352:1071–1080). The results from a second large Celecoxib polyp prevention trial are yet to be published, but apparently did not reveal any difference in thrombotic events between the placebo and the 400-mg per day treated group.What does all of this mean and what impact will these clinical results have? First, very few drugs have been examined over such a long time period with such a robust design and enough study participants to detect these kinds of side effects. Scientists must now question how many drugs currently on the market would stand up to this kind of scrutiny with regard to safety? Direct to consumer marketing of many new drugs has left the public with a false assurance of complete safety, when, in reality, all drugs have some side effects that need to be taken into consideration. Often, despite the rapid-fire disclaimers on TV commercials, the public does not understand that the true risks involved with new products are unknown until they have been studied in larger cohorts over several years. Most side effects become more apparent after a longer duration of therapy at higher doses of drug (ie, Viagra and blindness). Second, it appears that the safety profile of COX-2 inhibitors varies from drug to drug. The safety concerns of Rofecoxib were so high that the drug was removed from the market immediately. However, use of Celecoxib at doses of 200 or 400 mg daily seems to be fairly well tolerated. Of course, we await the results of the second Celecoxib trial before we can fully evaluate all of the safety issues. Finally, it has become apparent that all drugs approved for public use need to undergo a more robust post marketing safety evaluation, perhaps before and definitely after the first TV commercial lauding their benefits to mankind go on the air. About 12 years ago, my laboratory published an article in Gastroenterology documenting increased expression of COX-2 in colorectal cancers, without changes in COX-1 levels when comparing normal matched colonic tissue with colorectal cancers or adenomas (Gastroenterology 1994;107:1183–1188). Our initial observation was followed by several reports from other laboratories that confirmed increased COX-2 expression in colorectal cancers. These results were preceded by many observational studies indicating a decreased risk of colorectal cancer in people who consumed NSAIDs on a regular basis (Nat Rev Cancer 2001;1:11–21), although even today the precise dose and duration of NSAID use required for optimal risk reduction is debated. Most NSAIDs inhibit both the COX-1 and COX-2 enzymes, with COX-2 thought to be the primary anti-inflammatory target. The connection between COX-2 and colorectal cancer led many scientists and clinicians to investigate whether inhibition of this pathway could provide some reduction in risk for colorectal and other cancers. In the early 1990s, the pharmaceutical industry began developing agents that specifically inhibited the COX-2 enzyme without affecting COX-1 activity. Although these COX-2 inhibitors were initially developed and approved for use in treating patients with arthritis or pain, their appearance opened up the possibility of evaluating these drugs for a role in cancer prevention as well. The first clinical trial on cancer evaluated the use of COX-2 inhibitors in patients with familial adenomatous polyposis (FAP). The clinical endpoint measured was polyp size and number. After 6 months of treatment, a significant reduction in polyp burden was observed in those FAP patients taking the highest dose of drug (N Engl J Med 2000;342:1946–1952). These results led to the FDA approval for use of Celecoxib (Celebrex) in patients with FAP. This study also led to the design of large trials studying subjects who develop sporadic polyps to determine whether COX-2 inhibitors could also prevent polyp recurrence in that group. At least 3 trials were designed and initiated. One trial evaluated the efficacy of Rofecoxib (Vioxx). The other 2 trials evaluated the efficacy of Celecoxib. The design of these trials required continuous treatment for 3 years, with a 2-year extension to evaluate drug safety. Investigators evaluated subjects who had previously undergone a complete colonoscopy with polypectomy and were randomized to treatment or placebo arms. Drug safety was carefully monitored because these trials were designed to observe a large contingent of study patients over a long period of time. Most of the arthritis and GI safety trials conducted previously were no longer than 12 months in duration. The results of the “cancer prevention” trials received an enormous amount of worldwide media attention mainly because of safety concerns. The first trial results to emerge involved the Rofecoxib study, which demonstrated an increased risk of thrombotic events after several months of treatment in patients on the drug (N Engl J Med 2005;352:1092–1102). This drug did show efficacy with regard to causing a reduction in polyp recurrence. However, the safety concerns led to the removal of Rofecoxib from the market and a landslide of litigation brought forward by people who were exposed to the drug and experienced a thrombotic or cardiovascular event. These side effects would not likely have been detected in a 12-month study, since there was no significant difference between the placebo and treated groups during that time frame. Next, the safety results from one of the Celecoxib trials emerged, which again demonstrated cardiovascular side effects after several months of treatment, especially in the high-dose group (N Engl J Med 2005;352:1071–1080). The results from a second large Celecoxib polyp prevention trial are yet to be published, but apparently did not reveal any difference in thrombotic events between the placebo and the 400-mg per day treated group. What does all of this mean and what impact will these clinical results have? First, very few drugs have been examined over such a long time period with such a robust design and enough study participants to detect these kinds of side effects. Scientists must now question how many drugs currently on the market would stand up to this kind of scrutiny with regard to safety? Direct to consumer marketing of many new drugs has left the public with a false assurance of complete safety, when, in reality, all drugs have some side effects that need to be taken into consideration. Often, despite the rapid-fire disclaimers on TV commercials, the public does not understand that the true risks involved with new products are unknown until they have been studied in larger cohorts over several years. Most side effects become more apparent after a longer duration of therapy at higher doses of drug (ie, Viagra and blindness). Second, it appears that the safety profile of COX-2 inhibitors varies from drug to drug. The safety concerns of Rofecoxib were so high that the drug was removed from the market immediately. However, use of Celecoxib at doses of 200 or 400 mg daily seems to be fairly well tolerated. Of course, we await the results of the second Celecoxib trial before we can fully evaluate all of the safety issues. Finally, it has become apparent that all drugs approved for public use need to undergo a more robust post marketing safety evaluation, perhaps before and definitely after the first TV commercial lauding their benefits to mankind go on the air.