VVUS Vivus – Counting Votes for the QNEXA FDA Advisory Committee

---

Update 2/19/12: Please read out updated analysis here that reflects the draft meeting roster and new information found in the FDA briefing documents released last Friday.

---

On February 22, 2012, the Endocrinologic and Metabolic Drugs Advisory Committee will meet to discuss new drug application (NDA) 22-580 for QNEXA (phentermine/topiramate), submitted by VIVUS (VVUS).

VVUS’ panel is different from the AMGN, ASTX, and NGSX panels earlier this month. In those panels, we felt our behavioral analysis was unnecessary because the clinical and statisical data clearly did not support approval, even before the briefing documents were released. But for VVUS, we know from past obesity drug panels that the committee decision will be difficult and the vote will be split. Therefore, it’s worthwhile to go beyond our normal behavioral analysis and attempt to predict the make-up/votes of the upcoming VVUS QNEXA panel *before* the roster and draft documents are released.

Who will serve on the panel?

First, we compiled all voters from the previous VVUS, ARNA, and OREX obesity drug panels. We list all voter names, voting outcome, expertise and links to bios in the google spreadsheet below.

Next, we eliminated the voters who are not eligible to serve on QNEXA’s panel because their term on the Endocrinologic and Metabolic Advisory Committee ended. These voters’s names are highlighted in red at the bottom of the list. The original VVUS vote was Yes: 6, No: 10. If we eliminate the six voters whose terms have ended, the original vote would have been Yes: 4, No: 6. That’s a much slimmer margin than before.

We then added one current member of the Endocrinologic and Metabolic Advisory Committee who did not serve on any of the three panels. All current sitting committee members’ names are highlighted in green.

Finally, we conducted a subjective evaluation of who we think could serve on the panel. The box next to each name is blue if we think they might serve on the panel and red if we are think they will not. We assume all sitting members of the Endocrinologic and Metabolic Advisory Committee will attend. We also assume people who have served on at least two of the obesity panels will be invited to serve on this panel (their past votes are highlighted in gray). Below is our rationale for including or excluding each individual:

• Marshall Balish – Yes. FDA only included one neurologist on the last QNEXA panel. It’s plausible FDA would like a second neurologist’s opinion on topiramate’s potential side effects of suicidal ideation and cognitive deficits, just like FDA included two neurologists on the last CONTRAVE panel.
• Vera Bittner – Yes. Sitting member of advisory committee.
• Erica Brittain – Yes. Sitting member of advisory committee.
• David Capuzzi – Yes. Sitting member of advisory committee.
• Melanie Coffin – Yes. Every panel needs a patient rep.
• Heidi Connoly – No. Expertise in echocardiography and valvulopathy is specific to LORQUESS’s mechanism of action.
• Janet Cragan – Yes, served on last panel. Expertise in birth defects is applicable to FORTRESS teratogenicity results.
• Pamela Douglas – No. Expertise in cardiovascular imaging is important to valvulopathy and therefore specific to LORQUESS’s mechanism of action.
• Eric Felner – Yes. Sitting member of panel.
• John Flack – Yes. FDA would like to evaluate any links between elevated heart beats per minute and MACE, so the panel could recruit more cardiovascular experts.
• Jacqueline Gardner – Yes. We assume FDA will invite her since she served on both the LORQUESS and CONTRAVE panels. Her epidemiological experience will be useful regarding the FORTRESS results. Formerly a sitting member of the Drug Safety and Risk Management Advisory Committee.
• Edward Gregg – Yes. Sitting member of panel.
• Susan Heckbert – Yes. She served on the previous QNEXA panel, as well as the CONTRAVE panel. Formerly a sitting member of the Drug Safety and Risk Management Advisory Committee.
• Ed Hendricks – Yes. He served on the previous QNEXA panel, as well as the CONTRAVE panel.
• William Hiatt – Yes, for the same cardiovascular reason as John Flack.
• Jules Hirsch – Yes. He was originally invited to serve on the first QNEXA panel, but could not make it. He also served on the CONTRAVE panel via telephone.
• Sanjay Kaul – Yes. He has served on all three obesity drug panels so far. He is a sitting member of the Cardiovascular and Renal Drugs Advisory Committee.
• Elaine Morrato – Yes. She served on the last QNEXA panel as well as the CONTRAVE panel. Her pediatric outcomes expertise is important given FORTRESS. She is a sitting member of the Drug Safety and Risk Management Advisory Committee.
• David Oakes – No. Erica Brittain is the sitting biostatistics expert for the Endocrinologic and Metabolic Drugs Advisory Committee. There are many epidemiologists on this panel already, so they do not need another biostatistician.
• Michael Rogawski – Yes. He served on the last QNEXA panel, as well as the CONTRAVE panel.
• Ellen Seely – Yes. Sitting member of panel.
• Jodi Segal – Yes. She served on the LORQUESS panel. Her epidemiology expertise will help interpret the FORTRESS results.
• Ida Spruill – Yes. Sitting member of panel.
• Abraham Thomas – Yes. Sitting member of panel.
• Lamont Weide – Yes. Sitting member of panel.

This leaves us with a 22 member panel. A few people will inevitably have a conflict and be unable to attend, so this is just about the right size for the pool of potential FDA panel members.

How will each person vote?

Our goal is to extrapolate how each person will vote based on the comments he/she made in previous panel meetings. First, let’s review what’s changed since QNEXA’s last panel. For a quick summary, listen to VVUS’ webcast at the JP Morgan Healthcare Conference.

• SEQUEL study 2 year data was published in AJCN. No real surprises, but in our opinion the study suffers from survivor bias. Those who did well in the first year are more likely to continue during the 2nd year, so the results may not be indicative of true real-world outcomes.
• VVUS submitted an analysis of the heart rate/MACE event outcomes. This data is not public, so we have to take the CFO’s word for it that there is no link between heart rate and MACE. QNEXA lowers blood pressure, in contrast to CONTRAVE which raises blood pressure, so that’s a good sign. But as we will see from some members comments below, that may not matter and a pre-approval MACE trial may still be required, a la Orexigen.
• VVUS released FORTRESS top-line results for birth defects (cleft lip) linked to topiramate. As the CFO stated, there is an increased risk of cleft lip, although the prevalence is very low. He then goes on to state that topiramate remains approved for epilepsy. Weight loss is a completely different indication however – the person years of exposure for such a large patient population will turn even a low prevalence into a relatively large absolute number, so the risk/benefit ratio changes dramatically.

Let’s make the case for each individual vote:

• Marshall Balish – Yes.

  Balish served on the CONTRAVE panel and spoke three times. First, Balish agreed with statements from fellow neurologist Rogawski, who said “So in my view, the labeling should make it clear about the risks of seizures and I think this is an important factor. I personally feel that physicians should have at their disposal as many tools as they can to help patients.” Balish followed up on Rogawski’s comments by asking about making a stronger label. The second time he spoke, Balish voted in favor of a pre-approval MACE trial. The third time he spoke, he contradicted his vote for the pre-approval MACE trial and voted for approval, going back to his first statement. If Balish continues to agree with Rogawski, then he will likely vote yes, since Rogawski voted yes for QNEXA during the first panel meeting (see Rogawski explanation below).

• Vera Bittner – No.

  Regarding a pre or post-approval MACE trial for CONTRAVE Bittner said, “I voted to have the study done prior to approval, with rationale very similar to Dr. Heckbert’s. Since we require such outcome studies, for example, for lipid-lowering drugs, as an example, CETP inhibitors, it is my feeling that the obesity drugs should have the same level of evidence before wide implementation, especially since there seemed to be a dissociation between the safety signal that was seen with sibutramine and the subsequent increase in endpoints in that the blood pressure rise did not necessarily predict the individuals who had the adverse outcomes. So I think that we do need the hard endpoints to make the appropriate risk-benefit decision.” If the blood pressure rise did not correlate with MACE outcomes in SCOUT, then QNEXA’s decrease in blood pressure does not assure lack of MACE. QNEXA’s sample size is too small to pick up on a true safety signal. Therefore, we feel Bittner will vote no and hold QNEXA to the same “hard endpoints” standard as CONTRAVE and the other drugs she mentions.

• Erica Brittain – No.

  Brittain’s vote is a bit of a mystery because she did not serve on any of the previous obesity drug panels. As a biostatistican, we think she will pick apart the resampling methods VVUS used in the FORTRESS study to explain away topiramate’s higher odds ratio for oral cleft (5.44!). The FDA statistical analysis usually differs from the sponsor’s statistical analysis, so we expect the safety signal will persist and Brittain will vote no as a result.

• David Capuzzi – No

  Capuzzi voted no on the last QNEXA panel. His main concern was with prescribing phentermine to patients with a history of cardiovascular disease. He mistakenly voted yes before changing his vote to no, attributing his mistake to panel inexperience. He thought the panel’s concerns summaraized by Burman would be addressed prior to approval. VVUS has not fully addressed all of those concerns becaues the FORTRESS study is ongoing. Therefore, if Capuzzi thinks like Bittner regarding pre-approval requirements for MACE or if he believes the teratogenicity spike is real, then we think he will stick with his original no vote.

• Melanie Coffin – Yes.

  The patient rep for an obesity panel will vote yes because of the enormous pent-up demand for an obesity drug.

• Janet Cragan – No.

  Cragan is an expert in birth defects. At the last QNEXA panel she said, “But I think for a woman who has a seizure disorder that topiramate’s the drug that really will control her seizures, the benefits probably outweigh the risks that might be out there. But I’m not so comfortable saying that’s the case for a drug that’s likely to be marketed for a very common exposure, where there are a log of reproductive-age women who will be likely to take it, and probably a large number of inadvertent pregnancy exposures, I think.”

  She went on to say in her remarks explaining her no vote, “But in the end, I couldn’t really justify widespread use with the reproductive outcomes concerns that we have. And as I listened to the panel members discuss the other adverse events, it actually raised my level of concern rather than lessening it. I think the situation where the only way we’re going to resolve the reproductive risks, if we can, is to have a large number of women take the drugs and see what happens, is the situation we’re in with human teratogenicity, often. And that’s really difficult, but that’s just the nature of the situation.”

  The teratogenecity signal is real. Although topiramate’s risk/benefit profile is favorable for seizures, Cragan is not comfortable with this risk/benefit profile in the broader obese population. We feel Cragan will continue to vote no.

• Eric Felner – Yes.

  Felner served on the LORQUESS and CONTRAVE panels, voting no both times. His statements indicate he is concerned more about the magnitude of efficacy and long-term nature of the efficacy than with the general safety issues. Although Felner voted no for LORQUESS and CONTRAVE, we believe he will vote yes for QNEXA because it is the most efficacious of the three drugs and weight loss is sustained through the two year SEQUEL data.

  Here are his comments from the LORQUESS panel: “But I think when it comes down to it, I really didn’t have a lot of issues with the risk. But obviously some other people here did, and it made me look at it a little more closely. And then it made me look back at the real benefit at least today of this drug. And I just didn’t see it as being that efficacious for even the group that was studied. And what bothered me I think a little more about it was looking at the second year of the data. When they didn’t maintain weight as we’ve seen in some 10 other drugs, they actually gained the weight back right after the year. And if they would’ve continued past two years, it may have been even right back up to baseline even on the drug. … But that was the real key for me, was that I had a lot of trouble really seeing the benefit at least today.”

  Here are his comments from the CONTRAVE panel: “This is similar to what I have said at least at a previous meeting, that I’m not overly concerned about the events or the risks that we see today at one year of studies. I think more time would, obviously, help that. But I’m actually more concerned about the efficacy. I don’t think there’s a tremendous amount of weight loss when you really look at this in all four of their studies, especially if you throw out the one study that involves the lifestyle or the behavior modification, which I think we call safely agree that very few people actually do that in real world. But I am actually not that disappointed in the overall vote, even though I voted against it, because I think of all of the drugs that we’ve looked at, this definitely had the lowest signal from an adverse event and had very similar efficacy. But I think another year of information would be very helpful to see if the weight either goes down even further or at least does not increase.”

• John Flack – Yes.

  At the CONTRAVE panel, Flack said, “I think, though, that my gut was, after listening to everything, that this was not sibutramine 2, and that the company was clean, straightforward, did an excellent job, and I believe coined themselves well here today. So I had a level of trust that you don’t always leave here with.”

  Flack will be comparing QNEXA with sibutramine and CONTRAVE. Flack’s CONTRAVE comments focused almost exclusively on blood pressure. Since QNEXA has a better blood pressure profile than CONTRAVE, we think Flack will vote yes.

• Jacqueline Gardner – No.

  Gardner is a challenging vote to call.

  On one hand, she voted no for LORQUESS. She said during the panel, “I’d like to ask Dr. Williams if he would talk to us a little bit more about the potential for the findings in breast cancer in mice. He made a good point about prolactin not being associated with mammary tumors. But then you also were talking about developmental changes, and I wonder if you could speak to us about the issue potential in utero exposure and effect on unborn children if mothers are taking Lorcaserin.”

  Regarding a proposed pregnancy registry, she said “Pregnancy registries are very difficult to establish, to maintain, they’re very expensive, and they’re generally perceived to be a vale of tears even by the people who do them professionally, so in the absence of a hypothesis you might want to rethink that suggestion.”

  On the other hand, she voted yes for CONTRAVE. She said during the panel, “I’m influenced by the fact that these compounds have been in wide use and have been, in fact, evaluated repeatedly for different indications by other panels and the FDA, and some of those in recent years.”

  Explaining her yes vote she said, “Once again, the preponderance of the existing experience, the promise of definitive studies and monitoring going forward, and, as Dr. Flack said, the use of the word “potential” on two counts.” She felt that since NB were already approved individually, that they would be safe together.

  Gardner was concerned about the safety of unborn children with LORQUESS. She felt the combination of two previous approved drugs demonstrated safety with CONTRAVE. Given her experience with pregnancy registries, we feel Gardner will give more weight to QNEXA’s teratogenicity signal than the fact that QNEXA is a fixed dose combination of two approved drugs and will vote no for QNEXA.

• Edward Gregg – No

  Another tough call. At the LORQUESS panel he said, “I felt the weight loss, though not great, was acceptable. I felt the profile of risk factor benefits was encouraging, actually. And there wasn’t any clear evidence actually that this was a dangerous drug. That said, there was enough doubt raised about some of the risks, lack of understanding in some of the mechanisms underlying those risks that forced me to take a harder look at what I saw as the crux of the benefit, which would be on the magnitude of the risk factors, not necessarily the spectrum but the magnitude of the risk factor benefits after you think about and take into account loss to followup and so on. And when I did it that way the risk to benefit just didn‟t quite pass that bar, so on balance I think the drug is promising but it’s not quite there yet in terms of showing the evidence.

  QNEXA is more effective than LORQUESS. There is no risk of valvulopathy like with LORQUESS. So Gregg could view QNEXA as better than LORQUESS – greater weight loss, less dangerous drug. However, we feel he will waver on the teratogenicity data. That will raise a doubt in his mind, similar to LORQUESS, which we feel will prevent him from voting in favor of the drug.

• Susan Heckbert – No.

  At the previous QNEXA panel she said, “I agree with the comments of Drs. Cragan, Proschan, and Rogawski that I’m concerned about the teratogenicity here even though we don’t have a definite answer. And I think my concerns are such that I would recommend that we eally need to know more about the teratogenicity before this drug goes on the market rather than finding it out after it’s on the market.”

  At the CONTRAVE panel she said, “I voted no. Although the drug met the efficacy criteria, there was a concerning safety signal that we discussed at length in the blood pressure and pulse data, and I feel that we need to evaluate the effect on clinical outcomes before the drug is marketed.

  The FORTRESS study provides more data on teratogenicity and there is indeed a signal there. Therefore we believe Heckbert will continute to vote no due to safety concerns.

• Ed Hendricks – Yes.

  Hendricks votes yes at the last QNEXA panel, saying “I think that Qnexa does meet the FDA efficacy thresholds. I think the sponsor did an outstanding job of managing several very difficult clinical trials, and did an outstanding job producing the data, and that the data does show that the safety issues in the target population, which are the obese patients, that the drug is reasonably safe and that we should approve it. I think it does fill a gap in our treatment spectrum. I think if the drug is disapproved, we’re going to send a very board message to the obese and the overweight, and that that will further drive them away from medical solutions to this problem to all the various quackery things that are out there. And so I hope the FDA doesn’t go just by the beans, as we discussed.”

  On the CONTRAVE panel he said, “The same for me. Ed Hendricks. I voted yes. I think it’s a useful drug. I think we should put it to work.”

  As a private practice physician, he is clearly in favor of providing doctors with options. We think he will continue to vote yes.

• William Hiatt – Yes. At the CONTRAVE panel he said:

  “The question which I think I’d like to return to throughout the day is a fairly fundamental one; that is, if there is risk, you’re more likely to see it when the absolute risk is higher in patients with underlying cardiovascular disease than the intended use population. The question really is going to come to whether that relative risk, if it’s present in high risk patients, translates to the same relative risk in low risk patients, because when you go to an absolute risk that’s extremely low, you’ll never see it. And, specifically, your proposed randomized trial will never see it, because you’re going to study low risk patients.”

  “And the last one is, if the risks are really unknown about suicide and cardiovascular morbidity and mortality, the guidance is silent on how much risk should be excluded, how are we to think about that?”

  “In contrast, memory impairment may be a risk for some patients and not others. So potentially, on the older spectrum, mild cognitive impairment, those kinds of patients may deserve further study if this is used more broadly, because they may be more at risk for a memory impairment than perhaps a young, healthy, 4-year-old.”

  “My reason for voting post-approval was a struggle, but I think to impose the preapproval would likely kill the development. And I’m just wrestling with what that message would be to the broader community and that if we’re going to say to every sponsor, “You need a 10,000-patient event-driven trial to even be considered for approval,” that may be a steep hill to climb.”

  “Yes, entirely based on the Phase 4 commitment. I think that weight loss is a surrogate in that the FDA’s guidance and introductions got it right, that it really should be a means to an end of improving the overall cardiovascular risk environment of that patient, ultimately with a demonstration that there is reduced morbidity and mortality from the obesity.”

  Hiatt will take a broader view of the risks compared to Flack’s narrow view of blood pressure. He considered the many risks of CONTRAVE and still voted yes. His comments seem to indicate he will vote in favor of QNEXA as long as a proper phase IV trial is conducted regarding cardiovascular safety.

• Jules Hirsch – Yes

  At the CONTRAVE panel, Hirsch said, “But more troubling than even that is that all of the drugs, at roughly 20 to 30 weeks, the weight goes down, along with placebo, and then the bottom is hit; by 40 or 50 weeks, we get to the bottom. And then with every drug, with rimonabant and sibutramine and phentermine and so on, the inevitable increase beings. It goes over a year or two or three, and the vast majority, if not all of the people, are back to their starting weight.”

  “It’s extraordinarily important to make a prediction about how long the drug will have any utility, along with a lifestyle program. If I were to tell you that you only have about 14 or 16 months’ worth, that makes me feel very differently about the utility of the drug than if you tell me, oh, no, we have three or four years’ worth at least in here before the drug effect, whatever that may be, is lost.”

  “Yes, I am most distressed about this particular adverse effect. We, years ago, did studies of heart rate variability, and it’s perfectly clear, when you study that, that small amounts of weight reduction increase parasympathetic tone and reduce sympathetic tone. This is probably an important feature of the advantage of losing weight.”

  “I voted no, and I did because of, first of all, the very marginal efficacy of the drug barely meeting the requirements and the lack of any information as to what the efficacy might be after the one year. My prediction, on the basis of previous drugs and so on, is that the efficacy will wane rather quickly. I think the greatest worry about adverse effect is not the things we talked about, but the biggest adverse effect is going to be another drug failure, which is very likely to be the outcome of this, and that was the main reason for my voting no.”

  Although Hirsch voted no for CONTRAVE, we feel the results of the SEQUEL trial, with greater weight loss efficacy maintained over two years, will sway him to vote yes for QNEXA. While Hirsch may have some concerns about the increased pulse rate, they will be offset by QNEXA’s decreased blood pressure.

• Sanjay Kaul – Yes

  On the last QNEXA panel Kaul said, “I voted yes. There’s a very fine line between a yes and a no vote, and thankfully, the FDA pays more attention to the discussion rather than just counting the beans. My yes vote comes with a lot of conditions. And I will not hold it against the sponsor if they interpret my yes vote as a no vote. First of all, it should only be approved for low to medium dose, not for the high dose, because all the safety signals appear to cluster in the high dose. The sponsor should be required to conduct a pharmacodynamic study in the short term addressing the concerns that were expressed earlier on during the discussion, focusing specifically on commonly used medications, including cardiovascular and over-the-counter medications. There should be a clinical outcome study designed to rule our cardiovascular risk. It should be implemented within three to six months. The protocol should be with the FDA within three months. And the study outcomes should be available to the FDA for assessment within three to five years. It should be a conditional approval. If they don’t meet these 2 conditions, the FDA should have the right to withdraw the approval. There is already a FADAA program that sort of enforces postmarketing requirement for assessing safety signals. I have no doubts about its effectiveness. But I think it concentrates the sponsor’s mind if we impose, moving forward, a preapproval requirement to evaluate cardiovascular risk. And I think the FDA should seriously consider that as an option. It’s already been mentioned what patients it should not be given to. There should be strict contraindications in patients with arrhythmic heart disease, severe ischemic heart disease, patients with hypertensive heart disease, including TI and stroke, and it should be contraindicated. And there should be a prominent warning mentioned for adverse effects, particularly the drug interactions that have already 18 been elucidated. So as I said, a very fine line between a yes and a no vote.

  On the LORQUESS panel he said, “I really didn‟t think that the current portfolio of evidence was sufficient enough to persuade me that a desirable benefit risk had been demonstrated.”

  On the CONTRAVE panel he said, “In my opinion, the drug does not yield the weight loss efficacy margin that allows an acceptable safety trade-off consideration, especially given the adverse blood pressure and heart rate signals.” And regarding a pre (A) or post-approval study (B), “I voted A. However, I could have been persuaded to vote for B if the weight loss efficacy had been more impressive.”

  It’s fairly clear that Kaul favors QNEXA efficacy to LORQUESS or CONTRAVE. We think he will stick with his yes vote and append the same conditions, which as he said, could be interepreted as a no vote.

• Elaine Morrato – No.

  At the QNEXA panel Morrato said, “I was just doing some back-of-the-envelope calculations. If we take some conservative market estimate as to how many women are going to be exposed to this, in the materials that they share there’s about 9 million bariatric surgery patients a year. There’s about 6 million on phentermine. If we just assume out of that, which is really the high risk population, that this is being 1 claimed to want to market against, let’s assume one million women. And let’s assume that we are not able to, in the real world, achieve any better or worse than what they saw in their clinical trials, which was about a 1 percent pregnancy rate, if I understand right. You know, we’re looking at 10,000 pregnancies in just this select population that we’re feeling that the benefit and risk makes sense. If the drug is really used as broadly as we think it might be, then you have a much larger exposure in numbers of pregnancies. And as one said, I think in the open hearing, no one wants to conduct a large public health experiment on the population. I think we need to have a better sense of really what the risk is or constrain the use of the product so it’s really among those patients where we know the benefit really does outweigh the risk.

  We feel the teratogenicity signal will cause Morrato to continue to vote no.

• Michael Rogawski – Yes.

  At the QNEXA panel, Rogawski said, “Yes. I just wanted to state what may be obvious, but maybe not. And that is that these teratogenicity studies are extremely difficult to do. They take a long period of time, and there’s no way that they can be done premarketing. They have to be done in a postmarketing fashion when large numbers of women are exposed. So I just wanted to emphasize that fact, that this wouldn’t necessarily be a reason not to approve the drug.”

  “So I think that in terms of balancing the risks and the benefits here, I came to my conclusion to vote in favor of approval because it’s clear that these two components, as well as the combination, are indeed going to be used by patients because they are available. It seems to me that the best use by patients would occur with the most information and with the proper labeling, the proper education, and so forth, as would be done if the sponsor was marketing the drug and presenting the drug combination to the public. So I think, overall, there’s a greater concern with respect to public safety if we have non-approval because that means that we don’t have the benefit of the additional information and education, risk mitigation strategies, and so forth, being presented to the public. So that’s the reason why I voted in favor of approval.”

  We don’t believe the FORTRESS results will sway Rogawski into changing his vote, since he believes teratogenicity data can really only be studied in a large-scale post-marketing study. We think he will stick with his previous yes vote.

• Ellen Seely – Yes.

  On the CONTRAVE panel Seely said, “I think it’s through the larger study that we’ll be able to find out better who that population is that’s going to get benefit that will outweigh risk. And that given that the risk appears to be low, at least at one year, that the best way that we’re going to answer this in a way that’s going to be feasible for the company to be able to continue to market the drugs, and other companies to do drug development, is to use an approach similar to what the FDA mentioned, that the approval be granted, given the provision that a larger study takes place and that interim analyses will occur, and that that will allow risk to be really mitigated and define the population who could potentially get benefit, as well as just continuing the one-year study forward so that we can see whether the benefit is lost or whether we maintain benefit or even gain benefit over time.”

  We think Seely will similarly vote yes on QNEXA and propose a large post-marketing study.

• Jodi Segal – Yes.

  At the LORQUESS panel Segal said, “Can I ask the FDA a question? Is it sometimes better to approve drugs even if you have to pull them later? Is there a certain number of drugs that you‟re just going to have to get rid of later?”

  “I voted yes. Not because I think it’s the most efficacious, wonderful drug, but I think it met its efficacy requirements. I think it met its requirements for valvular disease. I agree that the cancer risk feels a little unsettled still. The depression and the dissociative problems seem to be not an issue. I think in the post-approval period there would have to be very active surveillance of everyone on the drug, although I realize that‟s not possible. But very active surveillance, large post-approval trials to follow-up on these endpoints, and claims studies as recommended to look for the rarer outcomes.”

  QNEXA is more effective than LORQUESS. If Segal thinks efficacy has been established and general safety concerns can be taken care of in a comprehensive post-marketing study, with the drug perhaps being pulled later, then she will likely vote yes.

• Ida Spruill – Yes.

  At the CONTRAVE panel Spruill said, “I’m representing consumers, and I want you to hear me that when I say that I believe that the benefits outweigh the risks — but, also, sponsor has considered the fact, and FDA, as well — is that we don’t have a high risk population. And if we talk about where do we go from here and what we want this to do, and how do we really want to stop obesity, then I think we need to seriously consider — and I’d support the outcome study post-marketing, as well. So this was difficult for me, because I voted no before.”

  She ultimately voted in favor of the drug despite asking for a pre-approval study. We believe her position as consumer representative will continue to push her to vote towards approval.

• Abraham Thomas – Yes

  Thomas started out conservative on the first QNEXA panel, raising all kinds of safety issues. But he became more liberal in the LORQUESS and CONTRAVE votes, despite stating there was marginal efficacy for both drugs. In Thomas’ case, we feel QNEXA had the misfortune of being the first of the obesity drugs to face a panel. Had it come later, he would have voted in favor. Thus, we think he will change his vote on QNEXA to positive to maintain consistency with the LORQUESS and CONTRAVE votes.

• Lamont Weide – No

  At the QNEXA panel Weide said, “If, with a year’s trial, you have double the depression risk and you have some cardiovascular questions, I would like to see it extended. I would like to see the at-risk population be sicker, if you will, so that we can find out whether or not these safety concerns are going to be a major issue. I would agree, I am really sick of taking medicines off of the market after they’ve been on a year or two because we’ve identified something that we didn’t know about. And that really is some of what has given the FDA a reputation outside in the public.”

  At the CONTRAVE panel Weide said, “I just pray and wish someday that we will have something that gives us 10 percent weight loss, that gives us long-term results for our patients, because let’s face it, obesity is a long-term problem. It is not a short-term problem. That means that when we’re looking at all these drugs, these agents that are coming before us, the question is going to be, if these patients aren’t taking it short-term anymore and they’re taking it long-term, what is our responsibility to protect them from the long-term implications of taking these medications and what are the risks?”

  Weide will get his wish regarding a 10% weight loss with QNEXA. But he has consistently voted against the weight loss drugs because he is concerned about long-term safety. We believe he will latch onto the teratogenicity data and vote no as a result.

Where does this leave us?

At the end of this, the tally is Yes 13 No 9. But there are some major caveats to consider.

The people most likely not to be asked to serve on this QNEXA panel are non-EMDAC members who served on a single panel other than QNEXA. Presumably they had some sort of expertise that was specific to the issues pertaining to the other drugs that we could not identify in our screening process. These voters include Flack, Hiatt, Segal, Balish, and Hirsch (who missed QNEXA and teleconferenced into CONTRAVE). We model all five of these voters as Yes votes, so if they are not included that will swing the vote to Yes 8 No 9.

FDA will likely inject some fresh blood into the proceedings. For this panel, will FDA want more consideration of the heart rate/MACE results or the teratogenicity results? If it’s the former, then the overrepresentation of cardiovascular disease experts on this projected QNEXA panel makes sense (there were fewer cardiovascular disease experts on the previous panel). If it’s the latter, then FDA will probably invite more members of the drug safety and risk management panel who are not listed here. The cardiovascular experts tend to focus only on the cardiovascular issues and will probably vote yes; including specialists like Flack or Hiatt will stack the vote in favor of QNEXA. Drug safety and risk management experts, on the other hand, will focus on the high odds ratio for oral cleft and will want to see the full results of the FORTRESS outcome. For instance, a current DSRMAC member like William Cooper who studies adverse effects of drugs during pregancy would probably vote negatively due to the teratogenicity. Inviting one type of expertise comes at the expense of another because advisory committees have a somewhat limited size.

If the vote is close, absences and conflicts could swing the vote either way.

Finally, a positive vote will not necessarily lead to FDA approval. This happened with OREX. Not every yes vote can be considered supportive – we would count the votes of Kaul, Hiatt and Spruill as votes that are nominally yes, but FDA could interpret as no, especially if FDA and the company cannot agree on the scope of a post-marketing study.

Update 2/15/12: There’s one more caveat to consider. This analysis attempts to predict how people will vote based on their comments at the previous panels. But even though the CONTRAVE vote was Preapproval 8, Postapproval 11 for a CV outcome study, FDA still required OREX to conduct the study preapproval.

The panel members didn’t know about the later pre-approval requirement, so it is not reflected in their comments above. Now the question is how much will the upcoming panel members’ thought processes change and be influenced by OREX’s study? In the FDA presentation, how much will they push a preapproval study based on OREX precedent?

What differentiates QNEXA from CONTRAVE that will allow VVUS to conduct a CV *and* teratogenicity study post-approval? Phentermine stimulates the production of norepinephrine, burpropion inhibits neuronal uptake of norepinephrine, so both drugs can theoretically raise blood pressure through the same noradrenergic mechanism, even though that was not seen in the QNEXA trials. Ultimately, what prevents FDA from imposing the same CONTRAVE pre-approval requirement on QNEXA?

Here is a summary table of each person’s voting history dating back to 2008:

Name	Num Votes	Proportion Yes	Proportion No	Proportion Abstain	Num Informative Votes	Num Inverse Yes	Inverse Yes	Num Inverse No	Inverse No	Liberal-Conservative Index	Maverick-Conformist Index
Sanjay Kaul	20	0.5	0.5	0	18	9	0.272	9	0.6328	0.4308	-0.2555
Eric Felner	14	0.7143	0.2857	0	11	7	0.2457	4	0.5393	0.3495	-0.2022
Elaine Morrato	11	0.7273	0.2727	0	10	7	0.2303	3	0.5015	0.3181	-0.1762
Abraham Thomas	12	0.75	0.25	0	9	6	0.3513	3	0.5064	0.2863	-0.293
Lamont Weide	8	0.5	0.5	0	7	3	0.1653	4	0.4393	0.2784	-0.0955
David Capuzzi	7	0.4286	0.5714	0	6	2	0.2404	4	0.6275	0.4352	-0.2238
Jacqueline Gardner	6	0.5	0.5	0	6	3	0.2159	3	0.4239	0.2484	-0.1391
William Hiatt	5	0.8	0.2	0	5	4	0.3517	1	0.7273	0.4969	-0.3599
Edward Gregg	5	0.4	0.6	0	5	2	0.1326	3	0.3774	0.2292	-0.0456
Susan Heckbert	6	0.3333	0.6667	0	4	0	0	4	0.5976	0.4791	0.0145
Ida Spruill	4	0.5	0.5	0	4	2	0.2659	2	0.3875	0.1987	-0.1758
John Flack	5	0.6	0.4	0	4	2	0.3114	2	0.1856	-0.0075	-0.1584
Melanie Coffin	4	1	0	0	4	4	0.4253	0	0	-0.2188	-0.2111
Janet Cragan	3	0.6667	0.3333	0	3	2	0.1378	1	0.375	0.2253	-0.0499
Michael Rogawski	3	1	0	0	3	3	0.3528	0	0	-0.197	-0.1419
Ellen Seely	3	1	0	0	3	3	0.525	0	0	-0.2489	-0.3062
Jodi Segal	2	0.5	0.5	0	2	1	0.6429	1	0.0588	-0.2283	-0.4363
Ed Hendricks	2	1	0	0	2	2	0.4034	0	0	-0.2122	-0.1902
Vera Bittner	2	0.5	0.5	0	2	1	0.1818	1	0.65	0.4743	-0.1746
Marshall Balish	2	1	0	0	2	2	0.2583	0	0	-0.1685	-0.0518
Jules Hirsch	1	0	1	0	1	0	0	1	0.65	0.5291	-0.0013