VIOXX Cardiovascular Safety Data form the APPROVe Study

VIOXX Cardiovascular Safety Data form the APPROVe Study

The Adenomatous Polyp Prevention on Vioxx (APPROVe) trial was a multicenter, placebo-controlled, double-blinded clinical trial to determine the effect of three years of treatment with Vioxx (rofecoxib) 25 mgm on the recurrence of neoplastic polyps of the large bowel in patients with a history of colorectal adenomas. Only patients not anticipated needing chronic NSAID (Non-Steroidal Anti-Inflammatory Drug) (drugs like Vioxx, Celebrex, Motrin, Ibuprofen, etc…) therapy during the trial were allowed to enroll. Low-dose aspirin therapy was allowed in up to 20% of the study population. 2,586 patients (62% male; mean age 59 years) were randomized in a 1:1 ratio to either Vioxx 25 mgm or placebo (stratified by low-dose aspirin use). Colonoscopies were performed at one year and three years, or at time of study therapy discontinuation. Patients who completed three years on study therapy were eligible to participate in an off-drug one year study extension with scheduled colonoscopy at the end of the extension. The three year base study was scheduled to be completed in November 2004.
The primary efficacy endpoint of the study was the incidence of colorectal adenoma, as evaluated by a study pathologist. Safety was evaluated through the reporting of AEs (Adverse Experiences) as with all trials. All investigator-reported SAEs (Serious Adverse Events) that represented potential CV (Cardio-Vascular) thrombotic events (blood clots) were blindly adjudicated by an external adjudication committee. Safety parameters were monitored on a regular basis by an ESMB (External Safety Monitoring Board).
Based on preliminary analyses, there were 25 confirmed CV events in 3,315 patient years on placebo (0.75 events per 100 pt-yrs) and 45 confirmed CV events in 3,041 patient years on Vioxx (1.48 events per 100 pt-yrs). The relative risk of a confirmed CV event was 1.96 (95% CI 1.20, 3.19; p=0.007). The relative risk for an APTC (Anti-Platelet Trialists’ Collaboration) was 2.25 (95% CI 1.24, 4.08; p=0.008). The results primarily reflect imbalances in myocardial infarctions and ischemic cerebrovascular events (white stroke). Although these events reflect an increased relative risk over the entire study period, the results for the first 18 months did not show an increased relative risk of confirmed CV events. The increased relative risk was observed beginning after 18 months of treatment. No difference in overall mortality was observed between treatment groups.
Based on the increased relative risk for CV events (even though the absolute risk is very small) the ESMB notified MERCK of their recommendation that drug therapy be discontinued. Merck followed this ESMB recommendation. Colonoscopy data remain blinded.
Dr. Saponaro’s Interpretation of the Data
Merck is a very conservative company that is truly looking after patients best interests, if anything, to a fault. The VIGOR (Vioxx Gastrointestinal Outcome Research) trial in 2001 pointed to some potential CV problems with Vioxx, but it was not designed well to address this properly. The APPROVe trial wasn't the best design to look at the CV risk issue, but was better than VIGOR. There were 2,586 patients in the trial for about three years. Remember, the point of the study was to look for a change in colon polyps, not CV risk. The study actually showed no difference in mortality. That is, no additional subjects died in the Vioxx group. There were 45 CV events in 3,041 pt-yrs on Vioxx vs. 25 CV events in 3,315 pt-yrs on placebo. Remember, there are only 2,586 patients in the study. The higher number refers to patient-years. That is, one patient on Vioxx for three years equals three patient-years. Even with the results exaggerated using pt-yrs, there are only an excess of 20 CV events over placebo. The P value is significant, but the actual number of people in the study who had a CV event is extremely small. In fact, I feel the number is too small to make a meaningful conclusion.
I think there needs to be a Vioxx vs placebo study done in a group at high risk for CV events that lasts more than 18 months. If I were in charge of Merck, I would have announced the results of VIGOR and APPROVe in the media and to the FDA. I would have announced that there is a possibility that Vioxx could be bad for you especially after 18 months of use. Stress that the risk is a CV one and that people at high risk or concerned about CV disease should avoid Vioxx. I would also announce the beginning of a new trial designed to answer the question of whether Vioxx is safe (free of CV side effects) when consumed for more than 18 months. Using a high risk group would enable a relatively small sample size and ability to accumulate events quickly.
Instead, what we are left with is speculations that Vioxx may be unsafe if used for more than 18 months. This has led to generalized speculations of safety of all NSAIDs (COX-2 specific and first generation). Many patients are stopping their arthritis drugs in fear and experiencing the expected morbidity as expressed with pain and loss of function. Lawyers are advertising looking for people that have suffered a stroke of heart attack who claim they took a Vioxx; even if was only one pill! This is a shame. The only question about Vioxx safety was when used for longer than 18 months. And, it still remains a question today. It is not proven that Vioxx is unsafe. In fact, if the proper trial were conducted, I feel it would be proven that Vioxx is safe and that the conclusions otherwise were a statistical aberration.

Sincerely:

Joseph Saponaro, MD, DABIM, FACP, CPI, CCI, CCRI, CCRC, CCRP
Board Certified Internist, JPMC
Principal Investigator, DSI
Diplomat American Board of Internal Medicine
Fellow American College of Physicians
Certified Physician Investigator by the AAPP
Certified Clinical Investigator by the DIA
Certified Clinical Research Investigator by the ACRP
Certified Clinical Research Coordinator by the ACRP
Certified Clinical Research Professional by SoCRA
Member: The American College of Preventive Medicine

Return to Table of Contents