Your Vioxx Lawyer - Vioxx, Celebrex & Bextra FDA Transcript

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.

NIH and Investigator Presentation:
Celecoxib in Adenoma Prevention Trials:
The APC Trial (Prevention of Sporadic Colorectal Adenomas with Celecoxib)

DR. HAWK: My name is Ernie Hawk. I work at the National Cancer Institute, currently Office of Centers, Training and Resources. The study I will share with you was done while I was a member of the Division of Cancer Prevention, and it is the APC trial. The story I would like to share with you over the next 20 minutes or so--I will be followed by my colleague from M.D. Anderson who led the Pfizer-sponsored PreSAP trial--the story I would like to share with you really has three important components. That is, the data that are available today; the data that I can't share with you today because they are still emerging. These two trials that I will discuss now are still ongoing. Drug administration was halted in mid-December with the finding of the risk that I will share with you today, but the trials remain ongoing, looking at efficacy and other issues with regard to overall safety. Then, finally, one of context because while the discussion this morning and early afternoon is centered upon the usefulness of these agents in inflammation and pain, they have another very important potential indication in terms of cancer risk reduction, both in a preventive as well as a therapeutic context. And, I hope to bring a bit of that to your awareness.

Depicted here is the disease in which we are attempting to intervene. It is colorectal cancer. When looked at globally or within the United States, despite the availability, as we heard earlier, of effective approaches to this disease in terms of screening and risk modulation through things like polypectomy, we remain with a significant problem, with 145,000 new cases anticipated in 2005 and about 55,000 deaths, and obviously a much larger issue ii the worldwide scene. So, the National Cancer Instituted is devoted to not only extending available techniques but exploring new areas to combat this disease.

I am trained as a medical oncologist and, therefore, my focus of attention and my training was to the right side of this slide, that is cancer. But cancer, as with most diseases, is actually a process--if only we had the tools to be able to identify it. Depicted here is the process moving from normal mucosa in the intestine through a variety of stages, intermediate polyps, adenomas, to invasive disease, invasive cancer.

This process is time dependent, taking typically years in most settings, and already this process is becoming the focus not only of cancer itself, but the process of cancer development has become the focus of clinical screening and surgical intervention when adenomas are identified, that is, they are commonly removed on identification. Because we are understanding the molecular pathogenesis of the disease, it provides opportunities to not only address at a pharmacologic level cancer itself but potentially to address the development of cancer through targeting of a variety of the parameters that drive the process on a molecular level. COX-2 is one important target in this process.

Depicted on this slide is the talk I usually give over the course of about half an hour. So, I will summarize for you the really profound amount of data suggesting that non-steroidal anti-inflammatories and COX-2 selective inhibitors may be useful in terms of preventing and/or treating cancer. The data is most compelling in the intestine, particularly the large bowel, however, as you will see it extends to other organs as well. It is one of the reasons why the NCI has invested so heavily in this area and why we believe it still holds great potential to benefit patients living with cancer or at risk for cancer.

There are four lines of evidence here that I would like to share with you with, again, probably hundreds or thousands of studies underlying these points.

On a mechanistic level, non-steroidal anti-inflammatories and COX-2 selective inhibitors have been shown to induce apoptosis of neoplastic clones, to reduce angiogenesis in animal models, to inhibit proliferation and to stimulate immune surveillance of neoplastic cells--all things which should retard carcinogenesis.

In vivo, in the intestine alone there are more than a hundred animal studies now published, 90 percent of which roughly show profound benefits in terms of reducing intestinal carcinogenesis, as depicted by reductions in cancer incidence, multiplicity in these animal models, delays in time to progression, reductions in advanced characteristics of tumors.

In terms of epidemiology, there are now more than 35 studies--retrospective, prospective, nested case control studies--which pretty consistently, with the exception of two studies, show 30-40 percent reductions across the spectrum of intestinal neoplasia, that is, reductions in adenoma incidence, cancer incidence and cancer-associated mortality.

So, we believe, based on the observational animal and mechanism data, that changes in adenomas will ultimately in the longer term translate into improvements in later outcomes such as colon cancer incidence and mortality, at least with this class of drugs, again, because of the really profound database. The epidemiologic studies alone amount to several million individuals involved in those studies.

Finally, there are now three published randomized, controlled trials of aspirin in the peer reviewed literature that suggest 30-40 percent reductions in recurrent adenoma. They were designed very similar to the APC and PreSAP trials that I will share with you in greater detail.

So, based upon this abundance of literature with its great consistency, we believe that non-steroidal anti-inflammatories and/or coxibs may very well reduce the risk of colon cancer. Importantly, what we learn in the colon may very well extend to other organs as well.

There are similar sorts of evidence, not nearly with the volume nor the consistency necessarily but suggesting that COX-2 is a relevant target to carcinogenesis in a variety of other epithelial organs, and that these agents may very well reduce risk of cancer development and/or be useful in cancer therapy.

I will point out that already these agents are used not only with the hope of preventing or treating cancer but also in treating many important conditions in cancer patients, such as pain and inflammation. coxibs are particularly useful because they tend not to interfere with platelet function, an important parameter in cancer patients because so many of our other therapies actually suppress bone marrow production, and we are faced with the situation where, with thrombocytopenia, we need to try to identify agents that are useful in those populations for other indications.

I won't belabor this point. You already know the safety concerns with traditional NSAIDs that are established. The question is what others lie out there still to be discovered because, indeed, as you have heard several times this morning, we don't believe that there are the same sort of information databases that we have now with celecoxib and, as we heard earlier, with rofecoxib with traditional NSAIDs.

We embarked on this effort to explore non-steroidal anti-inflammatories and coxibs specifically back in the late '90s when the data based upon the relevance of COX-2 to cancer development became apparent with the growing body of data I summarized two slides previously. So, we joined a collaborative relationship, a clinical trial agreement, with Searle, Pharmacia, Pfizer--a migration of companies over time--to evaluate celecoxib in a cancer prevention setting based upon the lines of evidence summarized here.

Our first attempt to do that was in a very high risk situation, that is, patients with familial adenomatous polyposis. I won't belabor this point greatly. This is a surgical specimen in the upper left and an endoscopic view of the intestinal burden of precancerous polyps in individuals born with this germline condition with defect in the APC gene. It is a relatively rare condition but confers essentially 100 percent lifetime risk of cancer if not mitigated by surgery or other maneuvers. So, typically these patients are subjected to a variety of standard care procedures involving genetic screening, endoscopic screening, surgical prophylaxis--actually removal of all or part of the colorectum, as well as standard surveillance for any remaining segments. Despite that standard of care, these individuals, compared to age matched controls in a landmark study done at St. Marks, one of the leading institutions for care of these patients, had a three-fold increased risk of death, mostly from cancer.

So, this led us to do a trial of celecoxib in which we showed efficacy. At the moment it is the only approved pharmacologic adjunctive therapy for this condition. However, earlier randomized, controlled trials had documented solidex efficacy as well, although that is not an approved condition.

These are the data that led to that approval under the Subpart H guideline, with further definitive trials required and those are ongoing and planned. This is with 6-month intervention involving 83 patients in a differential randomization, 1:2:2. This is endoscopic parameters. This is worsening below the line. Here is endoscopic improvement. What we see here, focused on the colorectum, is no mean change in the placebo group; a slight improvement at the 100 mg twice a day dose; and a substantial improvement at 400 mg twice a day.

Importantly, as someone pointed out earlier, individual activities are probably important because, clearly, some patients respond quite dramatically even to lower doses of celecoxib, although clearly you have a more profound and robust improvement at the 400 mg twice a day dose. Just as an example of what was seen, this is a non-selected patient. This is before. This is after 6 months of exposure, only 6 months of exposure with reductions in the intestinal tumor burden.

Importantly, these folks are at risk for duodenal cancer as well and we assess that in a variety of ways and feel that there is a suggestion of benefit in the upper GI tract as well, but it is certainly something that requires subsequent confirmation because that was not statistically significant.

That trial then led us to consider the possibility that celecoxib, as with aspirin and other agents that have been tried for adenoma prevention, may be useful in adenoma prevention in a cohort at moderate risk due to prior sporadic adenomas.

So, the NCI and Pfizer-sponsored APC trial was initiated. It involved 2035 patients with prior sporadic adenomas who were randomized in a balanced manner to celecoxib 400 mg twice a day, the dose that was effective in FAP patients, administered over 36 months versus 200 mg twice a day, a dose that had previously not been interrogated in oncologic settings versus placebo. It was conducted with a baseline colonoscopy, a colonoscopy after 12 months and after 36 months, evidencing adenoma recurrence, with collection of all adenomas while on trial.

The study was a major effort and really I should note the dedication of both the practitioners in the study team but also the patients involved, involving 91 sites, English speaking, most of those in the United States but with participation in Canada, Australia and the U.K. Accrual began in late November and extended to March of 2002.

Well, the trial moved forward with careful monitoring by the standing data safety monitoring board, and was largely unchanged until September of this year when, following the Vioxx announcement, the data safety monitoring board convened and recommended the initiation of a dedicated effort. Previously safety was a specified secondary endpoint but not cardiovascular safety specifically. So, they recommended to the steering committee that we initiate a process of cardiovascular adjudication and analysis focused on CV serious adverse events. So, that was done by drawing upon the expertise of a group of cardiologists and statistical team that is outlined here, based at Brigham and Women's, with the clinical endpoint committee involving two individuals who conducted the adjudication process in a blinded manner, created a database specifically focused on cardiovascular risk, and handed that off to a cardiovascular review committee, again with representation from Brigham, University of Glasgow and Dr. Wittes doing the statistical analysis.

The process of that adjudication, which we think is terribly important in this sort of trial that didn't up front specify cardiovascular endpoints, involved three steps. First of all planing. The team put together standardized definitions, hierarchical analytic categorization scheme and a statistical analysis plan.

Next, the data were compiled, verified and adjudicated. All SAE forms were reviewed along with source documents. Sites were queried to supply supplemental data focused on cardiovascular events. The events were adjudicated in the prespecified manner and a database was created for those events, handed off to the analytic team who then obtained randomization codes and relevant baseline data and analyzed the data according to intent-to-treat principles, and presented the data back to the data safety monitoring board in December.

Now to move to the data which has just been published on-line within the last 24 hours, I believe, on The New England Journal of Medicine web site. This slide depicts the baseline characteristics of the patients involved in the APC trial split out by treatment arm. What we see is that randomization worked quite well in terms of distributing these factors: Age roughly 60 years of age was the mean. About 70 percent of the cohort was male. About half of them had a history of some form of cardiovascular event that, of course, mainly being represented by hypertension in approximately 40 percent; diabetes in about 10 percent. Importantly, aspirin use and lipid-lowering drug use in this cohort was on the order of 30 percent and was balanced across arms.

When we come to the hierarchical characterization of cardiovascular endpoints--I had a heart attack when I saw the earlier presentation and different numbers were presented, but I realize that they were presenting the death from cardiovascular causes or myocardial infarction or stroke. That is the third line on this slide, not the fourth line which is the one that the steering committee and the safety assessment team chose to focus upon, which includes cardiovascular death, myocardial infarction, stroke or heart failure because we feel these are all clinically relevant and important outcomes that could be considered together.

Although the events are quite infrequent in this 2000 patient, 3-year study we see a differential occurrence regardless really of the categorization you are looking at when you are looking across treatment arms, whether expressed as number or percentage of patients or the rate per 1000 patient-years, there is a consistent increase in risk moving from placebo to 200 BID to 400 BID.

I will make a point that these are expressed as hazard ratios, that is relative to placebo, and again with all the various categorizations, and this really moves from the hardest endpoint, cardiac-associated death at the top, down through progressively felt to be more subjective assessments of cardiovascular risk to the bottom where you are dealing with cardiovascular death, myocardial infarction, stroke, heart failure, angina or need for a cardiovascular procedure. You will notice that the risk decreases as you move toward a broader categorization of cardiovascular events. But when you focus more specifically--again, I will highlight the blue line, the one that is highlighted in the manuscript--we see a 2.3 increased risk at 200 mg twice a day compared to placebo and a 3.4 increased risk compared to placebo at 400 mg twice a day. I will note that the lower number of 2.3 percent in the 200 mg group is marginally statistically significant but clearly significant at most assessments in the higher dose group.

If we focus, instead of that sort of assessment, on death, we see that there was a difference, not statistically significant per se from cardiovascular causes perhaps, but that death from non-cardiovascular causes does not follow the same trend. Indeed, when we look at death from any cause, overall mortality, there is really no significant difference across these arms, with the 200 mg group and placebo being equivalent in this study.

Similar to the discussions we have heard earlier, we considered a variety of cardiovascular risk factors based upon baseline characteristics at this point, and we evaluated age, gender, CV risk factors, diabetes, aspirin use and use of lipid-lowering drug use at baseline. We saw no statistical evidence, assessed by interaction terms, looking at the risk factor and treatment to suggest a differential hazard by any of those baseline factors. Of course, the analyses are limited by few events and, therefore, limited power.

If we look at a time to event analysis, with the Y axis including all 2000 patients and 3 years of follow-up, we see relatively slow event rates. However, if we then change the Y axis to focus specifically on a probability up to 5 percent we see the diverging curves similar to what was seen previously with rofecoxib, but the divergence coming somewhere arguably around 12-14 months. Importantly, these are intent-to-treat analyses.

I want to conclude with just a couple of points. you have already heard alluded to by Dr. Furberg the possibility of compiling a larger set of data from NIH-sponsored trials. Indeed, we have been very busy over the last several months trying to get this data in shape for presentation here from these two dedicated trials. I will point out that although the PreSAP trial is specifically sponsored by Pfizer alone, they shared their data with us and the adjudication and analysis process I described was applied to both the NCI-sponsored APC trial as well as the PreSAP trial, and funded by the NCI.

So, that was our first effort at an across trials analysis. You will hear in a moment from Dr. Levin that the analyses from PreSAP are not completely mature yet so we have a plan of doing this across the two that we have done as well as four others that we know exist that are NIH-funded, and it is simply a matter of trying to do this in an expedient manner at this point. These are the six trials that we feel have long enough exposure. That is, these are defined by at least 2 years of exposure and we generally try to shoot for sizeable trials, all placebo-controlled, because we felt these would be informative to the question at hand.

I will point out that the last study down there, the NEI study, is very small but a very high risk cohort and, therefore, they state they have a significant number of events, on the order of 20 events in just 86 patients.

Finally, I have tried to highlight for you that these agents may very well have a unique set of contributions to make to patients living with cancer or at risk for cancer, and we believe that strongly holds true and needs further investigation.

This study, with the caveats mentioned earlier this morning--this is an unpublished study but has come to our attention recently because we have investigators interested in looking at traditional NSAIDs given now the cardiovascular risk that has been identified with coxibs. What we see in a cancer relevant population, that is, patients with oral cancer in a closed population-based nested case control study in Scandinavia, is that the risk in this small study, unpublished yet, may extend to other NSAIDs. I think this, combined with some of the other observational data and the experimental data from the National Institute's of Aging study, may very well raise questions about other NSAIDs, and we think it is terribly important to answer those questions given the potential opportunity thee agents present for patients with cancer.

I will close with this slide, just stating that with most good research you are left with more questions than answers. Indeed, I think that is the case here. We believe that there are many issues still to be answered about this cardiovascular risk and what it means for patients with or at risk for cancer. I will leave this really to Dr. Levin to come back to at the conclusion of his talk, and turn it over to him at this point.

Weitz & Luxenberg is no longer accepting Vioxx cases.

see also:

Q & A Your Vioxx Lawyer - Vioxx, Celebrex & Bextra FDA Transcript - Q&A
Your Vioxx lawyer provides the complete transcript of the FDA meeting.

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

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