Industry Round Table: Where Are the New Drugs? (IAGS 2004)

Philip Reid (Eli Lilly & Company): Speaking as a cardiologist, not as a neurologist, I find that the challenge with the trial that Renato has just described is to demonstrate the efficacy of abciximab in acute nonhemorrhagic stroke patients. We must now try to transition what we know about the coronary arteries and the microvascular circulation to the brain. In a way, we are lucky, because thus far, it looks like the same pathophysiologic mechanisms that we belive or know take place in the coronary vascular system also seem to occur in the cerebral vasculature. As Renato also mentioned, we would hope that in the future, the pharmacologic therapy we use in the coronary arteries could be combined with some of the thrombus extraction devices that are currently being tested, and subsequently expand their application to stenting in the carotid arteries. Renato DeRita (Centocor): I will be attending the Stroke meeting next week in San Diego. When I scanned the list of participants submitting work for this meeting, I saw about half a dozen abstracts on the so-called “combolytic” approach to the same problem, in which a low-dose lytic is combined with a IIb/IIIa antagonist — either ReoPro or Integrilin, or even Aggrastat. I know several investigators who are trying that approach with varying degrees of success in their interventional work. There appears to be a really “dicey” reperfusion situation in which the operator is “opening the fire hose” in carotid stenting procedures where flow is restored very quickly. The brain is very unforgiving compared to the coronary arteries, so once flow has been restored, blood pressure must be carefully controlled. We are really just at the beginning stages of this type of strategy, and a good deal of work has yet to be completed. Richard Heuser: I work at a stroke center, St. Joseph’s in Phoenix, and I have to say that it would be very difficult for our center to conduct a study because the drugs will indeed be administered. An areteriogram is performed regardless of whether the patient is a responder or is suspected of being a nonresponder. And if any thrombus is found, the physician will likely take the “kitchen sink” approach. I do think it is important to do this, and I have emphasized this point at combined cardiology and neurology research meetings at my center. I have told them, “We are using the kitchen sink approach with this device, but we should go one step at a time.” How will this be accomplished, because many stroke centers will perform an arteriogram on these patients. It seems to me that this should be done at centers that don’t have 24-hour availability of a neuroradiologist to perform an angiogram and then a rescue procedure. How are we going to deal with that? Renato DeRita (Centocor): This is not an interventional trial. Richard Heuser: Yes, I know. Renato DeRita (Centocor): The centers participating are those that participated in the first study. It will indeed be a challenge to study the 600 patients who will be included in the companion arm. There certainly will be some “heavy lifting”, but we think it is important to do it. Johnson & Johnson has committed to seeing this through, and we anticipate about a two-and-a-half year time frame for the study. Philip Reid (Eli Lilly & Company): I think Richard made a good point because it’s almost a philosophical-medical decision. We don’t know which is better, so it comes down to the operator’s discretion. The majority of physicians, however, in the smaller, less sophisticated hospitals, would typically use medical rather than interventional therapy. The majority of stroke patients end up at these smaller hospitals. I suspect that Richard’s hospital is a tertiary referral center, which are important of course, but we are really aiming our intended medical therapy for the average to above-average type of hospital where pharmacologic therapy would be used first. We already know from the data provided from the t-PA studies that only about 4% of the eligible patients in the 0–3 hour time window receive t-PA. Thus, we face the obstacle of educating the physicians who will be using these new therapies, as they are not accustomed to administering lytics or antiplatelet agents. John Spencer (Genentech): I have been involved with stroke therapy and t-PA from the very beginning. It has been an uphill battle for us as a company to get the neurologists, who are not used to being on-call, on board with this approach. However, just this past year, we have observed a significant uptick in the amount of stroke therapies used. We estimate that this increase has been nearly 20% in 2003 for Genentech’s lytic business on the stroke side. Several major centers are administering lytic agents in combination with interventional neurological procedures. Some terrific results have been observed in these cases. In addition, the AHA has made patient education one of their primary goals this year, because most patients don’t seek treatment within the three-hour window of time. The thought leaders in neurology all seem to agree that lytic therapy is beneficial if administered within the three-hour time frame, and if longer than three hours, perhaps a lower dose should be used and the patient should be sent for intervention, or some combination of these approaches. Mike Cowley: Philip, what’s new at Eli Lilly? Philip Reid (Eli Lilly & Company): I would like to specifically address the major topics that would cause one to ask: “Where are the new drugs?” The odds of a medicinal chemist taking a thought and putting it on paper, creating a molecule that is a new chemical entity, not a knock-off of an older model, are quite small. The chances are 1 in 10,000 that this new drug will end up in the drug store ten years later. So right off the bat, companies like ours are struggling against the odds. I do think, however, that the future looks bright with some of the new intracellular techniques being developed such as DNA and gene splicing. We should see some bigger and better percentages of hits in the coming years. Today, however, is challenging. It takes about ten years before a new chemical entity arrives in the drug store. The device business is much different in terms of time; new devices are introduced about every twelve to seventeen months — and seventeen months versus ten years is a long time if you’re waiting for something new to come down the pike. The cost of bringing a new drug to market is currently about $1.5 billion, so not many companies can afford this investment. There are several factors involved in the long time frame required to bring a new drug to market. One is simply getting the study up and running. The complicated series of “hoops” to jump through, as you investigators well know, in order to launch a study are staggering. From the time we “shake hands and do the deal” to the time that patient enrollment begins is a minimum of six months, and is more often eighteen months. Furthermore, the rules have been changed by the IRBs and by the regulatory agencies in Europe. In the U.S., the HIPPA regulations are quite challenging. Obtaining a patient’s informed consent is increasingly difficult, and in a stroke study, this challenge is even greater and creates delays. The patient enrollment phenomenon is another issue we face. At the outset, investigators tend to be very optimistic about meeting their patient enrollment goals. Then, we encounter what we call the “hockey stick” effect in which the 45º diagonal of the time versus the number predicts what should have been accomplished; enrollment is typically slow at first and then it picks up later. Finally, another human factor comes into play in that we tend to be overly optimistic. We have our department check the milestone payments. Milestone payments are triggered by the enrollment of “x” number of anticipated patients, then a check arrives in the mail. When we looked at our records from 2003, 80% of the trial’s investigators missed their milestones. Thus, part of the drug delay issue involves the investigators. And then of course there are the outcomes of the trials themselves, which raises the question of why most drugs fail. Most of the time, failure is due to safety issues. Either we did not think of something, or some unforeseen issue arose. One example involves abciximab in the EPIC trial in which there was a 14% bleeding rate among the 2,100 patients, and it wasn’t until two studies later — PROLOG and EPILOG — that we discovered that heparin needed to be weight-adjusted. The bleeding rate dropped to 0.7% in the most recent trial as a result of this important information. Much progress has been made, so I don’t want to pretend that the drug development business is in a sad state of affairs, but the barriers are higher and it takes longer for a drug to reach the market. Scott Byrd (Eli Lilly & Company): As the commercial guy here on the panel, I would like to try to answer the same question that Philip was addressing, which is: “Where are the new drugs?” Naturally, I look at the issue from the market standpoint and from the physicians’ standpoint, and how these two aspects may influence what we are trying to accomplish on the “bench”. There are a couple of underlying drivers of what products are coming to market, or perhaps more accurately, what is not coming to market in the next couple of years. This Hawaii IAGS meeting is the first one attended by pharmaceutical companies. Your peers — interventionalists, surgeons, neuroradiologists, and so on — tend to be very excited about new device technologies, and for very good reason, because it’s an incredibly dynamic marketplace and has been especially so over the past couple of years. On the flip side of that coin is pharmatherapy, and now the combination of the two technologies — both delivery devices and pharmacologic agents. This cannot be overshadowed, however, by what is occurring in terms of the event rates as we are defining them in the marketplace. The standard of care has evolved dramatically in the last ten years in the field of interventional cardiology such that they have dropped to the 2% range. Also, improving care beyond the 2% event rate renders orders of magnitude more difficult with the next new therapy. The question thus becomes: “What is important to the marketplace?” What do we need to bring to market that can be commercialized and ultimately increase our business, while at the same time addressing the unmet needs? Indeed, even the way event rates are defined is changing. Historically, we have looked at hard endpoints such as death, as well as the semi-hard endpoints such as myocardial infarction (MI). Now, for the most part — even at the bigger meetings like the AHA and ACC — although re-MI is discussed on the podium as an important endpoint, it’s hardly ever discussed in practice. Most of our customers consider getting the patient off the table and achieving procedural success as the most critically important endpoint. In fact, the definition of MI takes on an almost laisser-faire approach, in which “CK leak” is the term that has replaced MI. This approach begins to influence what types of products come to market. Over the last couple of years, there has been a huge movement toward cost reduction as a primary goal for new interventional therapies, as opposed to improving clinical endpoints and clinical efficacy. This has affected not only who is conducting the trials — such as the smaller biotech companies like Millenium and The Medicines Company— but also the types of trials being conducted. The major interventional trials of the past two years have primarily been noninferiority trials, with a percentage of efficacy compared to the 100% baseline efficacy, yet at a lower cost. I am not belittleing that approach at all, because it can prove to be very successful and can deliver on an important need for the healthcare system. But what I am trying to highlight here — particularly at this meeting which focuses on advancing patient care — is that we are now entering into a new world in terms of what our highest priorities are. Those priorities, as evidenced by the actions taken by the marketplace versus what is said on the podiums, are what will drive these new markets. I think we will continue to see greater incentives for those companies that can conduct smaller trials and invent products that are nearly as good technologically, but are significantly more cost-effective. This will be the predominant driver of new therapies and devices in the coming few years. There are a couple of exceptions to this trend, which is why I’m glad we began by discussing the stroke trial, because it’s such an unmet need. These new major unmet needs will garner a lot of attention from big pharmaceutical companies who will develop therapies that will be able to penetrate the big markets. The industry is now comprised of Pfizer, Glaxo-Smith Kline, and perhaps, soon, a merged Sanofi and Aventis. These $20 to $50 billion companies need $2 billion drugs in order to grow. It’s extremely difficult to find a $2 billion-sized market, particularly in the smaller, acute therapy areas. Thus, you may see more biotech companies emerging as the primary deliverers of new technologies for the less common events. Having said that, there is a carryover effect. I know that Philip and some of my colleagues at Eli Lilly have been working on a new ADP inhibitor which will be investigated in our registration trial in the interventional setting. Thus, we are using the beachhead of high event rates in the cardiology arena and some of the more aggressive care to show that these products can be studied in smaller settings. But frankly, I don’t think that a trial would be planned if it weren’t for the $6 billion market that’s beyond the interventional setting. Much of what was discussed here earlier represents another exception, that of gene therapy, stem cell therapy, and so forth. Again, we must look at where some of that research will be done, and until there is more certainty involving the intellectual property protection issue, and limiting its risk, you will likely see more startup-based biotech companies venturing into those domains as opposed to big pharma. I wanted to make these comments to provoke this audience to ask: “What is our biggest priority?” — because that’s what will drive my colleagues to bring new products to market for your area. If the priority is cost containment, then that’s what you’ll get. If, instead, the priority is determined to be the advancement the standard of care, then we need to consider the types of trials that are currently being encouraged by industry. John Spencer (Genentech): I want to discuss what Genentech is doing in the area of targeted therapies. In the cardiovascular arena, we are looking at strategies for MI patients. When Bill O’Neill was presenting on the topic of acute MI the other day, I was quite surprised to see that there was really no discussion from this interventional group on facilitating approaches. There was always a “mantra” at Genentech that we would “fight the interventional guys”. It was “us versus them” and it took quite some time for me to convince my colleagues that this wasn’t an either/or thing and that we needed to partner with interventional cardiology to develop strategies that would work. One of the slides Bill O’Neill showed earlier involved a meta-analysis by Bates on trials of direct angioplasty versus lytic therapy. The analysis showed that there is a point at which the balloon is shown to be far superior to lytic therapy, and that crossing point where this was evident was at about 60 minutes. One might say 90 minutes, but beyond that, the national registry involving a huge patient population shows that the average patient gets to the catheterization lab and receives balloon therapy within 120 minutes or more. Many clinicians claim that nearly all of their patients get to the catheterization lab and are treated within 60 minutes or less, but I truly don’t ever see that. Thus, we have decided that a facilitated approach using a lytic agent in these patients is the best approach. We all agree at Genentech that lytic therapy is inferior to balloon therapy for patients who get to the catheterization lab within 60 minutes, but once that window of time has passed, we believe that lytic therapy offers greater benefit. This is particularly true for hospitals that aren’t equipped with catheterization laboratories. Genentech is funding a $100,000 program designed and launched by Michael Mooney. It involves employing a facilitated approach for patients who cannot get to the catheterization lab within the 60-minute time frame. I would like to hear from this audience about whether you think time does matter in terms of getting patients to the lab, or can patients be transported over two or three hours between institutions — or should we employ a facilitated approach for these patients? Genentech is spending a lot of money on the ASSENT 4 trial, which compares direct angioplasty with an anticipated delay of > 60 minutes versus lytic therapy, and transport to the catheterization lab fairly quickly. We have confidence in the benefits of a full dose of a lytic agent as opposed to a half-dose, combined with a GP IIb/IIIa inhibitor. Do any of you think that the facilitated approach using a lytic agent is beneficial, or are these two mutually exclusive? Mike Cowley: The real issue with lytic agents, in my view, involves patients with intracranial bleeding, lower efficacy in the patients who are coming from further out, and re-occlusion. But these are issues that could potentially be dealt with. If patients who are most likely to have intracranial hemorrhage are excluded (those > 70 years of age), there is not much downside for those patients Renato DeRita (Centocor): Given the discussion yesterday about the SEAPORT Trial, if I were having a big MI, I would want to receive what I consider the optimal therapy: to be brought immediately to a catheterization lab staffed with a skilled team, given ReoPro and a stent, possibly a drug-eluting stent (DES). I wouldn’t want to end up in a catheterization lab with inexperienced operators who are merely “dabbling” in angioplasty. The rub, however, is that such ideal conditions are not available to every patient, and certainly not in a timely enough manner. It’s sometimes difficult to get patients in to the catheterization lab quickly, even at Mount Sinai in Manhattan due to traffic congestion and other logistical problems. To help solve this problem, Mount Sinai has developed a strategy using a facilitated approach to buy time until the patient can undergo intervention if needed — because even when flow is restored, the underlying lesion still needs to be addressed. We are testing this facilitated approach in the FINESSE Trial. Working with the FDA to get the trial launched was not an easy feat. Many of the interventionalists were clamoring for the study to commence, but patient enrollment has been painfully slow, even glacial! Part of this is due to what Roxana Mehran touched on the other day — and that is the dearth of leadership on the cardiology side, not only in terms of relations with emergency departments, but with the community hospitals that are sending us these patients. The awareness simply is not there. The group I work with has spent a lot of time visiting outlying hospitals to educate their physicians about the facilitated strategy, why the trial is being conducted, who they should be in touch with at the tertiary center, and so on. Philip Reid (Eli Lilly & Company): In response to what Mike just said regarding safety, when we looked at the incidence of intracranial hemorrhage among the GUSTO 4 patients, age was indeed a critical factor. The majority of the intracranial hemorrhages and major bleeds occurred in patients who were > 70 years of age. We have gradually changed that rule to > 75 years, thus patients who are > 75 years of age receive a reduced dose of 75% of a lytic agent, which is fairly empiric. The older patients have strokes that bleed; that’s a simple fact. The second comment I want to make regards the facilitated PCI approach that John spoke about. I think that these strategies should be tested clinically, but I’m not sure that I consider them mutually exclusive. Instead, I would say that the technique must be evaluated in today’s practice. Facilitated PCI addresses an entirely different aspect of medical therapy delivery — the so-called “spoke and hub” concept. If a small outlying hospital wants to send a tertiary center a patient, there should be something the physician can do for the patient while preparing him or her for transport to the larger center. I am hopeful that trials such as FINESSE will be successful and not show that one approach is better than the other, but rather that the most appropriate medical strategy will be applied to each patient. Paul Overlie: I would maintain that a lot of what you’re saying, Phil, is already happening de facto. We conducted a t-PA versus primary angioplasty study years ago. I work in a region that has a fairly rapid patient transport system that goes out as far as 250 miles. At the same time, Genentech has their representatives traveling throughout our region to teach physicians at the outlying hospitals how to employ these therapies. I initially balked at that strategy, and then embraced it after a couple of years. My concern was always to get the patients to the lab as quickly as possible. John Spencer (Genentech): I agree, but a small shift recently occurred when the DYNAMI data were released. I wish Bill were here — at many meetings, I have heard him say that time really doesn’t matter, because beyond two hours, the potential for salvage is negligible. DYNAMI was misrepresented; in that trial, the lytic arm versus the primary angioplasty arm showed that the time difference between the arms was 50 minutes! None of us here would disagree that if treatment is provided within Kenneth Kent: Our center has adopted the facilitated angioplasty strategy for the past two years. We treat many acute myocardial infarction patients, 80–90% of whom are sent to our facility from outlying hospitals. When we receive a call from these smaller hospitals, we instruct the treating physician to administer thrombolytics at one-half or full dose, according to that physician’s discretion, and then have the patient transferred to our center. Again, in the absence of any definitive studies, and even going against some of the studies published in the literature, this facilitated approach has worked very well. The biggest question still involves whether GP IIb/IIIa inhibitors fit in to this therapy. Again, the difference between the ADMIRAL and CADILLAC studies confused everyone. The one difference between those two studies was the time at which the agent was administered. The second point I want to discuss involves the variable comfort levels of the physicians who administer these agents. Our staff members have gone out to speak with the referring physicians at the hospitals located 50–100 miles from our center and educated them about administering lytic agents, but obstacles arise such as a new partner or a physician covering their practice who isn’t familiar with this strategy and is reluctant to employ it — even if we limit lytic agent administration to patients ? 75 years of age. The studies will help, but I hope they will include the question about whether GP IIb/IIIa inhibitors are useful in that patient population. And once those studies are completed, an enormous educational campaign will be necessary. Philip Reid (Eli Lilly & Company): I think there is a complementary action between lytic and GP IIb/IIIa agents. Unfortunately, GUSTO 4 didn’t enlighten us much in that regard. One aspect worth discussing is the ambulance systems, which in most parts of the U.S. are fairly closed. We have been working with several ambulance companies that serve one central hospital instead of dropping the patients off at the nearest emergency department. They have the IRB-approved authority to administer GP IIb/IIIa inhibitor agents, but have been reluctant to acquire the right to do so. Thus, a third strategy for managing STEMI patients involves administering a GP IIb/IIIa inhibitor in the ambulance or, if the patient is a walk-in MI, the patient may be given thrombolytics and a GP IIb/IIIa inhibitor, depending on the hospital’s protocol. I consider PCI to be a fundamental therapy. Kirk Garratt: In the interest of time, I would like to steer this conversation toward the topic of new drugs being developed by the companies represented here. One area of interest to me is immune modulators. I know that Genentech has been researching white blood cell inhibitors and similar molecules. According to what I’ve seen in the data, white blood cell inhibitor therapy to improve outcomes post-MI has been a total failure; it has only served to increase infection rates. I would also like to hear from the companies here about apoptosis-inhibiting or modulating agents that are potentially applicable to both MI and stroke. John Spencer (Genentech): Genentech’s CD-18 failed miserably because we thought that the microcircular vascular bed was being clogged. We are currently re-engineering our VEGF molecule which initially failed partly due to the patients who could not be revascularized. We tried to go the easy route with stem cell research, administering it IV followed by two IC courses of the VEGF, and many patients responded to the VEGF and grew new vessels. We have re-engineered the VEGF molecule to make it longer-acting. We are currently in phase-three trials. VEGF will probably initially be used for wound healing and bone healing, two areas in which it has shown incredible results. Genentech is not currently working on anything in the area of apoptosis. However, Genentech is developing a factor Xa inhibitor at the top of the cascade, but I can’t talk much about it at this time. We will have approval very soon for an anti-VEGF compound for solid tumors which could be worth about $2 billion for our company. Renato DeRita (Centocor): With respect to the early work Genentech did on white blood cell inhibition, it gets back to what David Holmes mentioned on day-one of this meeting, and that is that the therapy might work, but there are issues of timing and up-regulation of these receptors and cell pathways. If they aren’t hit just right, then the effect is not seen. We know this even with some of the drug-eluting stent work in the early days: if the right cell isn’t being targeted at the right time, both with the right dose and in the right tissue setting, then an effect will not be seen. And it’s not because the therapy doesn’t work, it’s just that it hasn’t been applied correctly. Centocor is a monoclonal antibody company, and the new therapies in our pipeline use these antibodies to treat inflammation, solid tumors and other conditions. We are also very involved in looking at ways to produce these molecules more cost-effectively, because we’ve been fighting the abciximab cost issue for a long time. The fact is, it takes three months to manufacture a batch of abciximab, a process which involves about 75 different chromatographic steps, so it’s very complicated and costly to manufacture. I was recently informed that Centocor made its first batch of monoclonal antibodies that were produced in corn. This is very exciting and will hopefully bring the cost of the technology down, perhaps not to the level of synthetic molecules that are built with reagents right off the shelf, but certainly considerably lower than we can currently produce them. Monoclonal antibody production currently uses mammalian cell culture techniques. Centocor is focusing on anti-inflammatory agents and some anti-cancer agents based on our monoclonal antibody therapy, as well as anticytokines which primarily target pro-inflammatory cytokines. Centocor is also active in tissue factor research and the techniques that will hopefully be used to generate less expensive quantities of clinical-grade material. Philip Reid (Eli Lilly & Company): The list of new products in the pipeline at Eli Lilly is not quite as long. We are working on a new ADP inhibitor which we hope will address one of the current clinical drawbacks of these inhibitors — Plavix resistance, for example, which can occur in up to 30% of patients. Our new ADP inhibitor should help reduce later morbidity in those patients. Eli Lilly is also working on a Xa inhibitor that looks very promising. We are also developing an activated protein-C, which Eli Lilly currently offers for infectious shock patients. Protein-C has been found to work in the microvascular system and the coronary arteries as well. Eli Lilly is considering conducting a trial that will combine abciximab plus this activated protein-C inhibitor in STEMI patients. With respect to the question of apoptosis, most of you know that there are distinct anti-inflammatory effects with abciximab; we are working on proving their clinical relevance. Among those effects are the inhibition of apoptosis and acutely ischemic myocardial cells. Adam Greenbaum: I would like to return to the topic of facilitated angioplasty by taking the devil’s advocate stance There have been instances in the past when things looked great on paper, but didn’t necessarily work out, such as inotropes for CHF, antiarrhythmic agents for arrhythmia, and so on. Likewise, facilitated angioplasty may actually end up being detrimental to patients in the long-term — we don’t know this yet. Currently, there are no data showing that opening the artery or increasing TIMI 3 flow at the time of angioplasty translates into improved mortality. We know from PACS that the arteries are open when you get there, but there is no change in the mortality rate. We know now from various single-center studies and multicenter meta-analyses that angioplasty is more effective than lytic agents. We used to think that the most effective treatment was to open the artery and remove the obstructive lesion — and that was in the days before distal protection was available. Perhaps the way reperfusion actually occurs with a lytic agent or even a combined approach is that there is more distal embolization as the artery is being opened, leading to greater microvascular damage. And it may turn out that doing nothing, then opening the artery using distal protection is better than a slow reperfusion that causes increased distal embolization. ADVANCE AMI has been on hold due to lack of enrollment, FINESSE is having difficulty, and ASSENT 4 is trying, but it may be that physicians are looking at these data and deciding that there’s no need to facilitate TIMI 3 flow if the artery is going to be opened up with a balloon using means to protect from distal embolization. Mike Cowley: The data on distal protection are scant at present also. Adam Greenbaum: Yes, that’s true. But it seems to me that many of you here are saying, “Why doesn’t everyone believe in the efficacy of facilitated angioplasty?” I just don’t think there are enough data to support that approach yet. Mike Cowley: Thank you everyone. We’ll finish this discussion on that provocative note!