Questioning the Vioxx drug - Are Valdecoxib and Parecoxib Safe for Treatment?

Your Vioxx lawyer provides you the complete transcript of the February 16^th, 2005 joint meeting of the FDA's Arthritis Advisory Committee and the Drug Safety and Risk Management Advisory Committee. We have formatted the complete transcript of the three day conference for easy of navigation to provide you with the best possible Vioxx information. To contact a Vioxx Lawyer, click here for a FREE case evaluation.

Sponsor Presentation:
Cardiovascular Safety and Risk/Benefit Assessment of Valdecoxib and Parecoxib

DR. VERBURG: Thank you very much. I will be brief. The next 25 minutes or so are focused on the cardiovascular safety of valdecoxib and the parenteral prodrug of valdecoxib, parecoxib, and brief risk/benefit assessment.

Just by way of some quick background, valdecoxib was approved in the U.S. for the indications of osteoarthritis and rheumatoid arthritis in November, 2001. The approved dose is 10 mg once daily. In terms of the NDA database, over 15,000 individuals were studied, which was roughly comparable to that which we supplied for celecoxib. Since the approval we have been focusing this drug in terms of its effects in acute pain as well as other non-arthritis chronic pain conditions.

Our overall assessment or position of valdecoxib is stated on this slide. First, it is our view that valdecoxib remains a viable treatment alternative for patients with osteoarthritis and rheumatoid arthritis. We have data to suggest that valdecoxib provides improved GI safety compared to NSAIDs. The valdecoxib cardiovascular safety database is smaller than celecoxib at present, however, the emerging CV safety profile of valdecoxib appears similar to alternative therapies in arthritis patients, that being non-steroidals, for up to 6 months. And, the cardiovascular signal in the CABG surgery setting, therefore, does not appear to extrapolate to the arthritis population based on the data at hand. Just as with celecoxib, we have shown that valdecoxib exhibits the properties expected of a selective COX-2 inhibitor. That is, as shown on the left-hand portion of the slide, it provides efficacy--in this casein a trial of osteoarthritis patients--that is superior to placebo treatment and in every way comparable to patients treated with naproxen at a full therapeutic dose. At the same time, over the same 12-week period we see that the rate of endoscopic ulcers, with valdecoxib doses ranging from 5 mg to 20 mg once daily, were similar to placebo and in contrast, again, to the results seen with the non-selective agent naproxen.

In our prespecified, predefined way, we also evaluated 8 of the randomized, controlled trials with durations of 12-16 weeks in the NDA database, and also evaluated the same in 3 open-label studies up to 1 year. This was done according to prespecified definition, prespecified protocol and by a blinded events committee.

What we see here is that the incidence of ulcer complications, that being GI bleeding, obstructions and perforations, were significantly higher in the combined NSAID group, that being comprised of naproxen, ibuprofen and diclofenac, as compared to placebo treatment. No such difference was seen in the valdecoxib treatments as compare to placebo at doses ranging from 5 mg up to 80 mg daily. We also see that in long-term exposure at doses of 10 mg to 80 mg daily out to 1 year in nearly 3000 patients the event rate seen with valdecoxib looks similar to that which we saw in the more short-term but controlled settings.

As I have mentioned before, there are more limited safety data than with celecoxib and the analysis is largely confined to the randomized, controlled trials in arthritis at present, as well as some short-term acute pain studies alone or valdecoxib in combination with parecoxib. There are no completed epidemiology studies to report, although we are aware of three that are ongoing.

In the meta-analysis of valdecoxib there were 19 randomized, controlled trials included, with a total of over 12,000 patients. Again, the majority of the patients were osteoarthritis and rheumatoid arthritis patients, with a smaller minority of patients with chronic low back pain or chronic cancer pain.

The distribution of patients is shown here. The study duration ranged from 2 weeks to 12 months, and 11 of the 19 studies were 3 months or longer in duration. We evaluated all doses of valdecoxib in the meta-analysis. In terms of exposure, 50 percent of the patients treated with valdecoxib were exposed to the drug for periods of 3 months or longer; 22 percent for 6 months or longer; and 4 percent for 1 year or longer.

Here we show the distribution of events, as well as the event rate, comparing valdecoxib at doses of 10 mg or higher, in other words, it is full therapeutic dose and super-therapeutic doses that were tested as compared to a combined NSAID category. We find here that for the composite endpoint of cardiovascular death, non-fatal MI or stroke valdecoxib compares with a somewhat lower rate than that seen with the NSAIDs. There are very few events to shape this comparison, 21 in total. But as we go down to the various components of that composite endpoint we see essentially the same kind of pattern. If we translate that into a relative risk comparing valdecoxib to NSAIDs, we see that the point estimates of relative risk all lie under 1, or the large confidence intervals do not allow any strong conclusions to be drawn at this point.

On this slide we break down the comparison of valdecoxib to the individual NSAIDs, as well as compare it to placebo. Again, this is in terms of a composite of these 3 events. We see for placebo that the risk estimate is 1.26 but not significantly different. Then breaking out the comparison between naproxen and diclofenac, we see that again the point estimates are below 1 but not significantly so. No comparison could be done against ibuprofen. There were no events in either the valdecoxib or ibuprofen patients in which to do so.

Again, we have very limited information to establish any type of dose-response relationship or relationship of dose to cardiovascular safety with valdecoxib. The data that we do have are shown here. At 10 mg and 20 mg the point estimate compared to the NSAIDs is below 1, not significantly so. We noticed the point estimate moves to favoring NSAIDs at a 40 mg dose, however, when we move up to 80 mg there were no events in either treatment group in which to shape a conclusion. Again we are moving to very small numbers of patients as we begin to subdivide the meta-analysis for valdecoxib.

In terms of the incidence of cardiorenal events, as was the case with celecoxib, there are significant differences in the incidence of adverse events--these are investigator reports now--as compared to placebo. But comparing valdecoxib at doses of 10 mg or greater to NSAIDs, we see that there are no significant differences for either hypertension, edema or cardiac failure in over 7000 patients in this particular evaluation.

There is one other safety issue that we need to bring up with valdecoxib, that being the reports of serious skin reactions. Spontaneous reports of serious skin reactions, that being Stevens-Johnson syndrome, etc., received approximately 6 months after the launch of valdecoxib in the U.S. This rate appears to be higher than celecoxib or rofecoxib and, as a result, a black box warning was added to the prescribing information for valdecoxib or Bextra in November of last year.

In brief then to summarize the conclusions, valdecoxib shows efficacy that is similar to NSAIDs, and there is emerging data to establish that GI safety benefit is superior to NSAIDs and the CV safety profile is comparable to NSAIDs.

The added warnings allow physicians to choose appropriately based on the evidence of rare although severe skin reactions. The future plans for valdecoxib are very similar to those proposed for celecoxib. Longer-term studies are planned to evaluate the GI safety and the cardiovascular safety of valdecoxib in the arthritis patient population.

Now briefly a discussion of parecoxib. Parecoxib is the water soluble prodrug of valdecoxib. Its water solubility allows it to be administered parenterally either by intravenous or intramuscular injection. Parecoxib itself does not have any inhibitory activity at the COX-2 enzyme but, once administered, it is rapidly converted to valdecoxib. In fact, there is nearly total conversion within 30 minutes following administration.

So, the choices of parenteral therapeutics for the treatment of acute pain, whether it be post-surgical or other conditions, are fairly limited. As a result, there is an additional need to provide agents that improve the postoperative pain control or other acute pain situations with parenteral therapy.

As we have seen from various reports, inadequate postoperative pain is one of the most important factors in prolonging hospitalization and also in progression of acute pain to chronic pain following surgery.

Postoperative analgesia at present is traditionally provided by opioids but we all are aware of the complications of those therapies, and also opioids do not provide adequate analgesia upon movement and, of course, both of these issues also prolong the post-surgery recovery course. There has been an increasing move towards the use of multimodal analgesics, that is, drugs from two or more classes, to minimize the adverse effects of the drugs alone by reducing the dose, or to improve the ultimate efficacy output. In terms of the addition of agents to opioid therapy, therapy are very limited at present for parenteral therapy and basically limited to ketololac which has issues of its own in the post-surgical setting but, nonetheless, when studies are done this allows for early oral intake, ambulation and hospital discharge. The net comment on this slide is in the box here, which is that parecoxib is intended to provide significant analgesia, while sparing opioids without the GI and bleeding risks of parenteral NSAIDs.

This is just some data that illustrates the point that I made on the previous slide. These are two studies taken from the ambulatory surgery setting. On the left is a study of nearly 4000 patients who were asked to evaluate their pain one day after surgery, and we can see here that over 25 percent still had moderate to severe pain despite being treated with standard of care opioids.

Over on the right-hand portion of the slide is a much smaller study, conducted in patients undergoing laparoscopic surgery or hernia repair, and what we can see is that patients struggle to return to their pre-surgical functional status after surgery. Although the time course of recovery is somewhat dependent on the type of surgery they undergo, there is still significant functional disability several days after ambulatory surgeries.

So, by way of background, parecoxib was approved for the short-term post-surgical pain in Europe in March of 2002. At this point in time over one million patients have been treated. The parecoxib NDA is currently under review in the U.S. for the management of acute pain.

In total, the clinical registration program for parecoxib looks as follows: There are 64 studies completed. Of these, 26 were analgesia studies. In total there were about 10,000 patients randomized to one of the three treatment groups shown here; 1670 patients received treatment for 3 or more days with parecoxib and over 1000 patients received treatment for 10 or more days with parecoxib and then transitioned to oral valdecoxib therapy.

One of the earlier studies that we performed in the program was in a high risk surgical population to gauge the overall safety of parecoxib. We chose the CABG surgery population to perform such an analysis. This was a 2-treatment arm, double-blind, randomized, controlled trial. Patients were randomized in a 2:1 fashion to active treatment. Following CABG surgery they received parecoxib at 40 mg IV Q 12 hours for a period of at least 3 days, and then once they were able to transition to oral treatment they received valdecoxib at the same dose.

The other treatment arm received placebo treatment throughout the course of the 14 days. Both treatment groups received as needed supplemental analgesia in the form of morphine during the parenteral treatment period or oral acetaminophen codeine during the oral treatment period.

It is important to note that prior to randomization or receiving study medication all patients were to receive 80-325 mg daily of aspirin. Approximately 89 percent of these patients underwent CABG surgery with cardiopulmonary bypass, and about 11 percent of the patients were off-pump cases.

Here we show the results that emerged from the first CABG surgery trial. We had put in place an endpoint committee to adjudicate the events in this trial according to prespecified definitions and, of course, they were blinded to study treatment. What we see here is that if you look at the composite endpoint made up of these various components, we see that over the course of the entire trial there was an increase in the incidence of any thromboembolic cardiovascular event in the parecoxib/valdecoxib treatment group. This result was driven primarily by this imbalance that we see here in stroke or TIAs, although those incidence rates and differences between the treatment groups did not achieve statistical significance.

In light of those results, we elected to evaluate the cardiovascular safety of parecoxib and valdecoxib in a larger CABG surgery study. The study design is outlined on this slide. In total, this was a trial of over 1500 patients. Following CABG surgery the patients were randomized to one of three treatment groups. They either received parecoxib as a 40 mg IV loading dose and then 20 mg IV Q 12 thereafter, transitioned to valdecoxib after a period of 3 days, and then for an additional 7 days of oral treatment.

The second treatment group received placebo during the parenteral period and then was transitioned to valdecoxib during the oral treatment period. The final treatment group received placebo throughout both the parenteral and oral treatment periods. Again, patients were able to receive parenteral supplemental analgesia as required. As in the previous CABG trial, all patients were to receive aspirin at doses of 75-325 mg daily per protocol. In this trial all CABG cases were performed with cardiopulmonary bypass.

We used a slightly different adjudication scheme in this trial as compared to the first trial, and we were focused in this trial on myocardial events, cerebrovascular events, peripheral vascular events and pulmonary embolism.

Here we show the results of the CABG surgery trial in terms of the overall composite endpoint of all adjudicated thromboembolic cardiovascular events. This is broken down to the intravenous, oral and entire study period. If we look at the entire study period which, by the way, included not only the 10 days of treatment but also a 30-day post-surgery follow-up period, we see that parecoxib/valdecoxib was associated with a significantly higher incidence of thromboembolic cardiovascular events as compared to patients who received only placebo. Patients who received only valdecoxib had a numerically higher incidence of thromboembolic cardiovascular events. This difference did not achieve statistical significance. Also, as you scan up here you can see that actually 3 of the events in this treatment arm occurred in patients prior to the point that they ever received valdecoxib.

Similar to the results seen in the evaluation of the crude incidence rates, here we show the time to event analysis for the parecoxib/valdecoxib treatment group, valdecoxib only and the placebo group, again, out to 30 days post last dose of study medication, as stipulated per protocol. Again, we see that based on log rank test the parecoxib/valdecoxib treatment group was significantly different from the placebo group. No significant differences were seen between the placebo and the valdecoxib only treatment groups.

If we now break down the composite of cardiovascular events into its components and quickly look at the various components, we see again, as we saw in 035, that a major driver for the difference overall is the CVA or TIA category, as well as cardiac arrest and cardiovascular death which tended to occur later in the treatment of parecoxib/valdecoxib, while the strokes were clustered quite closely to the post-surgical setting.

By the way, I should probably state that as a result of those findings we quickly went to those countries, those regions of the world where parecoxib is currently marketed and have modified the product labeling in those areas to contraindicate parecoxib and valdecoxib in CABG surgery or in other revascularization procedures since those types of settings have not been studied.

We have also taken the step in the U.S. of including a contraindication for the use of Bextra of valdecoxib in the CABG surgery setting or revascularization setting even though Bextra does not carry an indication for acute pain.

I just want to go back to the CABG surgery setting and make some concluding remarks. Again we are faced with limited data to really evaluate the effects of parecoxib and valdecoxib as compared to NSAIDs in this treatment setting. There is very little data with respect to NSAIDs.

The mechanism for the increased cardiovascular risks with parecoxib and valdecoxib is not known. We did risk factor analyses but, as you can appreciate with the small number of events, that didn't prove to be too fruitful.

We do know that patients that undergo CABG with coronary bypass pump result in activation platelets, leukocytes and endothelial cells; that aortic cross-clamping results in ischemia, re-perfusion injury and emboli formation. There is a complex time course of changes in prostacyclin and thromboxane-A2 that have been reported following CABG surgery. And, as Dr. FitzGerald mentioned this morning, this patient population is also characterized by a high degree of platelet aspirin resistance. So, the constellation of all these factors obviously in some manner contributed to the results that were observed with parecoxib and valdecoxib, but the importance of all of these factors in that respect cannot be sorted out with the current study. What we do know though is that some of these are isolated exclusive to the CABG surgery setting.

At the same time we conducted a study in CABG surgery patients, we also undertook a study in general surgery patients. This was basically an all-comers trial. Only patients undergoing transplant surgery, intracranial surgery, revascularization procedures or partial liver resections were excluded from the trial.

The doses tested and the duration of the trial are very similar to the CABG trial. The same endpoint committee was employed. Events were adjudicated in the same manner, according to the same definitions as the CABG trial. This was a 2-arm trial evaluating parecoxib followed by valdecoxib versus placebo and, as in the previous trials, patients could receive additional analgesic medication as required.

So, if we look at the incidence of adjudicated thromboembolic events in the general surgery trial, we see that the event rates--these were crude event rates--were one percent in the placebo group and one percent in the parecoxib/valdecoxib treatment group. This study also had a 10-day treatment window as well as a 30-day follow-up period. Again, we see that the distribution of events is scattered through the components of the composite with no clear patterns established.

The time to event analysis is shown here. Again, no differences were seen in this analysis by log rank test.

We have also expanded our evaluation of the cardiovascular safety of parecoxib to all the surgical trials that we have performed with this drug. Here we are showing such an analysis, excluding such minor surgeries as third molar extraction, etc. We are really focused here on the more complicated surgeries, whether they be orthopedic, etc. We had about 1900 patients in the placebo group; over 2600 in the parecoxib treatment group. Again, we saw no differences in the incidence rates. We tried to collect as much information as we could over the entire parecoxib registration database. Very quickly, just a brief word on the benefit that we see with parecoxib. I want to turn to the general surgery trial, first showing you the analgesic results that were observed with parecoxib and valdecoxib in this trial. We saw significant reductions in pain across the entire treatment period with parecoxib/valdecoxib as compared to standard of care alone. In fact, these reductions were fairly impressive. They were on the order of 25 percent or more. Those improvements in analgesic efficacy came in the face of significant reductions in overall morphine or opioid requirements to control pain. There was a 35 percent reduction overall in the use of requirement of morphine across the trial in the parecoxib/valdecoxib treatment group as compared to placebo. You can see that most of that effect occurred during the parenteral treatment period. With that also came an improvement in opioid-type side effects but also, perhaps as importantly, it also came with improvements in functional status of the patients following surgery. Here we show the Modified Brief Pain Inventory Functional Questionnaire, and you can see that there is a significant improvement in function in the parecoxib/valdecoxib treatment groups as compared to patients who received standard of care opioids only.

Finally to sum the risk/benefit of parecoxib, parecoxib appears to offer unique benefits over existing parenteral analgesic medications and has a favorable risk/benefit ratio in surgical settings, other than CABG or revascularization procedures. Because parecoxib is a parenteral, it is administered in controlled settings under physician observation. This risk/benefit assessment is also shaped by the cardiovascular risk that is found in the CABG surgery setting but no in other surgical settings. Again, the caveat is that we have no evaluated the drug in other revascularization procedures and have no assessment of safety in that regard.

At this time I would like to turn the podium back over to Dr. Feczko for some concluding remarks.

Weitz & Luxenberg is no longer accepting Vioxx cases.

see also:

Valdecoxib and Naproxen The threat behind Valdecoxib and Naproxen - What Vioxx committee says
Valdecoxib and Naproxen can be potentially harmful in Vioxx - Read why

Cardio Events & COX-2 Your Vioxx Lawyer - Vioxx, Celebrex & Bextra FDA Transcript - Cardiovascular Events, COX-2
Your Vioxx lawyer provides the complete transcript of the FDA meeting.

FDA Transcript - Day 1 Conference for Safety - Vioxx, Celebrex & Bextra FDA Transcript
Read here for the complete transcript of the FDA meeting for Vioxx