February 26, 2025
Welcome to Perspectives, a signature podcast series from The Leerink Center for Pharmacoeconomics. Hosted by Dr. Mel Whittington, a health economist and Head of the Center for Pharmacoeconomics, we will be hearing from individuals across the industry to better understand and appreciate the societal impact of healthcare innovations.
Mel Whittington: Hi, everyone. I’m Mel Whittington. I’m a pharmacoeconomist, which means I’ve spent my career building economic models to examine the costs and consequences of pharmaceuticals to inform conversations around things like efficiency and value. And today I’m joined by Dr. Frank David, who spent his career also working with financial models, working with investors and innovators to examine investment and development opportunities in biotech. So pharmacoeconomics, which is my world, and then the healthcare investment and development space, which is Frank’s world, historically haven’t overlapped very much, which is pretty wild given drug pricing is so important to both of those worlds. So today, Frank and I are going to try to overlap. We’re going to talk about the forecasting and modeling that are core to both of our respective worlds and talk about how we approach decision making. And I imagine we’re going to find out that it’s probably pretty different between both of those worlds. And we’ll leave it up to you all to figure out if that’s a good thing or a bad thing, or if it’s just a thing. So, for the pharmacoeconomists listening, I hope this conversation gives you a little glimpse into the healthcare investing and development world. And for the investors and innovators listening, I hope this gives you a little glimpse into how the field of pharmacoeconomics might operate. So, Frank, thank you for being willing to have this nerdy conversation with me about building models to support decision making.
Frank S. David, MD, PhD: Yep. Super happy to be here. Long time listener. First time caller, as they say.
Mel Whittington: I love it! So, first, can you introduce yourself? Tell us a little bit about what you do in the healthcare investment and development world and, you know, what drew you to it?
Frank S. David, MD, PhD: Sure. So, I started out training to be a physician scientist, did an MD PhD, did a residency, went back into the lab and then decided I was more interested actually in how drugs get made than in making data. So, I was a little bit at a loss at that point. I didn’t really know anything about how drugs got made and got some really good advice that I should check out life sciences consulting as a way to learn the space and potentially as a long-term career. And I actually started my career at Leerink which had at that time had a small consulting group that had grown out of MEDACorp. We had about 50 people. And we did everything in the life sciences space. So, we did everything from R&D strategy all the way to pre-launch planning to life cycle management. We had one partner at one point who did pricing and reimbursement and market access. So, it was a great exposure to really all of the kinds of questions that get answered. From the very earliest stages of a drug development product all the way to the end, genericization and, you know, long term life cycle questions. So, I spent five years there and it was great and decided that I wanted to see what it was like to be on the inside in a drug company. So, I went to AstraZeneca and did sort of similar work, but from the inside. So, working on R&D strategy, so really focused on the early pipeline and decided that actually I like consulting better. So, I left AZ in 2013 and formed Pharmagellan and never looked back.
Mel Whittington: I love it. So, as part of Pharmagellan and I believe you’ve authored two guidebooks. Is that correct?
Frank S. David, MD, PhD: Yeah, so I’ve always liked writing and that was one of the things that drew me to consulting. I just like the fact that a lot of the consulting deliverables are written, and I like the craft of writing. So, the first book that I wrote with two colleagues was really on biotech financial modeling and it grew out of some of the experiences from Leerink where and what we’ll talk about this a little bit more later, but we were often called to build models of these very early stage drug development programs and the problem at these early stages is you don’t really know a lot about the program and so you need some for lack of a better term, stub numbers to put into your model to try to do some sensitivities around and try to at least get a ballpark figure of what the program might be worth, and it was really hard to find those. And there were weird sources. We had internal documents that people had cobbled together on these PDFs that got shared around sort of furtively, but there was no authoritative book that you could go to where you could just quickly find inputs when you were in a data free zone for an early-stage biotech model. So, I thought it would be worth putting all of that stuff together. And that’s what we did. We just put together a book where we collated all of the best information that we could find on what kind of inputs to use when you’re truly in a data free zone building an early-stage biotech model. And then later on, after that, wrote a book about how to interpret clinical trials. Again, most of the people that I work with are interested in strategy, but they’re not the people who are designing and executing clinical trials. They’re mostly the consumers of clinical trials, whether it’s a commercial group or a senior executive type of person or whether it’s an investor. So, for them, what I had realized through the consulting was that it was important to help them understand the clinical data, not to turn them into statisticians or clinical development experts, but just so that they could be conversant and know where the bodies are buried, so to speak. So, the second book was about interpreting clinical trials.
Mel Whittington: Yeah. So, I’ve recently bought both of those books. I’ve read the first one, the biotech forecasting evaluation. I finished it in one sitting. It was this book that I’ve just been looking for and it was there, and it existed. And I loved it. And this is jumping ahead a little bit, but I think it kind of alludes to the importance of this conversation. In my world of pharmacoeconomics, we do not typically think of genericization or loss of exclusivity or how prices change over the course of our model. And our field has really just started beginning thinking about, is this necessary? Do we need to do this? Are we able to do this? Do we have the data to be able do this? And one of the rationales for not doing so and rather doing what’s more conventional of keeping that price static over time, is uncertainty about what inputs to use. To your point like, well, what do you use? How long is the exclusivity period? How do prices change over the exclusivity period? What will the price do after the exclusivity period? And so that’s something that the world of pharmacoeconomics is actively grappling with and figuring out like is there any evidence to be able to inform this? And then, you know, I’m reading your book, and I see all these recommended inputs and citations from literature as supporting evidence as to why those are good inputs to use. And I’m like, “Oh my gosh, when was this book published? Is this hot off the press? Is this why like these inputs haven’t reached the pharmacoeconomic world? Like, am I just stumbled on something so new here?” And I, and I looked at when the book was published, and I think it’s like eight years old and I don’t think it hasn’t reached the pharmacoeconomic world because it’s brand new, because I think it is eight years old. It’s more that here’s this pharmacoeconomic world, and then there’s this healthcare investment and development world that have been operating very separately and haven’t been communicating.
Frank S. David, MD, PhD: Yeah. I mean, I think some of it just comes down to why people in my world build these models in the first place, right? So, I would say the classic situation is you have this early-stage asset, whether it’s your own asset or one that you’re looking to in license, for example, and you’re really trying to get a bead on what it’s worth. So, what it’s worth has basically four components. There’s during the R&D phase, there’s cost and time and risk, and then on the back end, there’s reward. And so, you’re really trying to figure out how those things all fit together, and the standard way to do this, from an investment point of view, whether you’re a financial investor or a strategic investor, like at a company, at a drug company, would be a risk adjusted NPV model. So, you’re basically trying to figure out the net present value of all of the future cash flows, both outflows and inflows, and you’re trying to account for all of the risk that happens along the way during the R&D period in particular, because obviously these trials are not all going to succeed and we’re very early in the process. So, what happens is that you know you’re going to be wrong, right? Every input by almost by definition is going to be wrong, but the value of building the model is to force you to articulate your assumptions and figure out which of the different levers are actually going to have the biggest effect on the risk adjusted NPV at the end of the day. So, that’s why we need these stub numbers and these assumptions, not because we believe that is absolutely the right number, but because then we can have a conversation about let’s say phase three probability of success. So, here’s a value from the literature on what the average probability of success is for oncology drugs. Let’s say or for colon cancer drugs more specifically and then we can say “Okay, based on what we know about the science of our particular drug and other factors, do we think we are par above par or below par and what kind of arguments would we make to convince ourselves that one way or the other, and by how much, and then what is the impact of those assumptions if we play them out and look at the impact on the risk adjusted NPV.” So, what you often see people do is they’ll build a model and then they’ll end up with one of these kind of tornado charts where they go through a bunch of the key inputs and for each one, they’re basically using some midpoint assumption as the middle of the tornado, and then they’re having, they’re flexing each input between a high estimate and a low estimate and trying to figure out what’s the impact on the NPV and what you might end up finding, for example, is that what really is affecting your model quite a bit is the cost of the phase three trial, for example. Above a certain point it becomes NPV negative or sort of marginal, but if we can reduce it to X amount, then it becomes positive, and so that becomes interesting. And I think specifically to your point about the genericization, because it’s an NPV model, those very far out years when you’re talking about like a phase one drug, for example, don’t have a huge impact on the NPV, but their impact is nonzero, certainly. And again, the point is not to exactly know how much value is there by being on the market for eight years versus nine years versus ten years, but it’s to force you to articulate what do you think is the right answer to how long it’s gonna go before it goes generic and then to have a discussion about how sure do we do we need to be about that right now in order to make a decision about this project?
Mel Whittington: Yeah, there’s so many things that you said that I find fascinating. So, I would also say in the models that I build, these pharmacoeconomic models, it’s also kind of to figure out what’s it worth? What are the health and non-health outcomes that you get from this drug? And then is it worth the price that is set by the market? Or by the manufacturer and so very similar premise of why we create these models is really to inform resource allocation and efficiency and in pharmacoeconomics we also do tornado diagrams. Tornado diagrams are also my favorite thing to put into reports because there’s so much that can be captured there, right? But we change each of our model inputs across upper and lower bound and then see what it does in pharmacoeconomics or in cost effectiveness analyses what you’re really looking at is the incremental cost effectiveness ratio. And so how does that change with your inputs? And so, in my world, the inputs are typically, the treatment effect, what’s the hazard ratio on survival or a rate ratio of exacerbations or, quality of life estimate, and then be able to see across these various values, these plausible values for these inputs, does your cost effectiveness estimate change to the point where your decision would change and then it kind of helps you figure out are we pretty certain about that input or are we uncertain? And how does that percolate through to our findings? One other thing that I think we’ll get to in a little bit, but I want to bring it up here because you mentioned about, the genericization and being far out in time but, but that far out in time is you looking at maybe a phase one drug, and so it’s really far out in time and pharmacoeconomic modeling and cost effectiveness analyses, these are often done at launch. And so, you know, genericization isn’t as out as it is. And so, here’s a situation where we’re closer to genericization, and we’re not including it as compared to you guys being further away.
Frank S. David, MD, PhD: Yeah, and obviously, when you’re building a model for something that’s near launch or has already is peri-launch. It’s a whole other ballgame because there you’re really trying to understand what your PNL is going to look like in terms of the running the business. So, then it really does matter how many more years. It’s not so much about net present value. It’s about year to year, how much money do we think we’re going to deliver to the business? And they are obviously thinking about the genericization and when that’s going to happen is critically important because how much are we promising to the CFO and for how long is a major piece of running a franchise at a big pharma company. So that’s a place in time, like with your models, where really having a good bead on what you think is going to happen in those five, six, seven, eight, ten years out from launch is critically important. I’m not saying they’re not important when you’re looking at phase one, but all other things being equal, I would say that the financial impact on a risk adjusted NPV model is somewhat less, than it is, it’s more of a strategic question at that point. So, when you think about, for example, some of the debate about in biologics versus small molecules under the IRA and the fact that biologics have always been favored, right, because the biosimilars market is essentially not highly functional, let’s just say, whereas the generic small molecules market is pretty highly functional. So, there’s always been a penalty, but now that penalty under the IRA is exacerbated. I would say that from what I’ve seen, that gets incorporated into more of a strategic discussion than a model per se. The impact on the model is less than the impact on the, than the overall corporate strategy. Because you kind of know that sure, those out years don’t matter now, but if we’re lucky enough to launch a drug, those out years are gonna matter a lot in terms of delivering revenues to the business. So, we don’t really care that it doesn’t impact the NPV model that much today because we know from a strategic point of view, we would like to be more heavily weighted toward biologics and small molecules. So, I think when you think about the IRA and sort of how it impacts decision making in that particular case, I would say it’s a little bit more on the strategic and qualitative level than it is on the quantitative level.
Mel Whittington: Understood. Going back a little bit about to just why we create models. Your book started out with the quote, “all models are wrong, but some are useful” by George Box. I think that quote was probably on 50 percent of my lecture slides. The quote works from the pharmacoeconomics world as well, so I’m not going to argue with the quote. I think it’s true, all models are wrong, but to not have them, I think you’re more wrong. And in the foreword of your book, it says, “models make poor decisions based on incorrect assumptions less likely.” And I was like, yes, that’s what we’re trying to do, right? We’re trying to be a little less wrong. But you know, they do give us a template or a roadmap to, as you say, line out our assumptions. Set up our logic and to inform decision making and a little bit more of a structured way. You talked about NPV models a little bit. I want to contrast those with models that I typically build and then you know fill in any gaps that that you want to add but in my world of modeling we typically do maybe, often cohort models for cost effectiveness analyses where really you’re tracking this hypothetical cohort of patients across these health states absolutely let’s say it’s Alzheimer’s disease, you start out, you know, a cohort of patients and you kind of track them through, mild cognitive impairment through mild Alzheimer’s disease, moderate Alzheimer’s disease, severe Alzheimer’s disease, death, and really model these two worlds, one in a world where there is the drug that’s being evaluated and one in a world where standard of care exists, whether that’s another drug or whether it’s a do nothing. And then you’re tracking the costs and the health outcomes over that patient’s time horizon, which is often their lifetime. And there’s nuances here, right? Like you can make a variety of different assumptions, you can take different perspectives, look at different time horizons, have different model health states and look at different things. But that’s kind of the, the basic concept of a model for cost effectiveness analyses is to track this population through different health states and see what are the costs that they get? What are the health outcomes they get? There’s also another type of model I should allude to and that’s budget impact models. And that really is more so looking at what does a new intervention, what’s its impact on an overall budget? That’s more focused on affordability rather than like, what is a drug worth based on the health and outcomes gain. So, I’m not going to, budget impact models aren’t the major point of discussion here, but I thought it’d be interesting for me to talk through kind of what is a cost effectiveness analysis and how it’s set up. Our inputs are not cost of goods sold or cost of trials or regulatory. It’s really more like health system costs, patient health benefits, survival, quality of life. And so just thought it’d be fun to contrast that.
Frank S. David, MD, PhD: Yeah, I mean, one place where they come together, I think, are some of the assumptions around what’s the nature of the patient population. Because if I’m building a model for an early-stage asset in Alzheimer’s disease, then a lot of the structure of the model and the inputs flows from questions around what population do I believe I’m actually targeting here? And how big is that population? And what other treatment options do those patients have besides slash in addition to my drug? Those are all going to affect the market size at the end, right? Because now I know, okay, there are all of these patients with Alzheimer’s disease, but actually my drug is mainly aimed at the patients with mild to moderate Alzheimer’s disease, for example, as opposed to severe, or as opposed to mild cognitive impairment. So, I need to have some assumptions around what the size of that population is. And more importantly, what it’s going to be in the future. Because I’m in phase one, maybe that population is going to change in some way due to differences in diagnostic procedures, differences in other treatments, et cetera. And I need to put some stakes in the ground in terms of what I believe that’s going to look like. Similarly, if I’m developing this early-stage Alzheimer’s drug, I need to have some view of both the current competitive landscape and what the future competitive landscape is going to be. So, I need to have some view of what are these patients going to be treated with? What are their options going to be besides or in addition to my drug? And that’s going to affect how much of the market I think I’m going to capture. And then these things also percolate backwards into what happens when you start thinking about the trials that are required to execute in these different areas. So, if I’m looking at a population that’s more severe and maybe I think that I’m gonna have a pretty modest effect on those patients, then that might translate into a certain size of clinical trial with a certain amount of cost and time associated with it, maybe a certain amount of risk as well. And those are gonna be different than if I’m looking at an early-stage population where I think I’m gonna meaningfully bend the curve in terms of how they progress, and it’s going to have a really great value proposition. And in fact, in companies that really do portfolio review in a formal way, which is mostly large pharma companies. That’s really baked into the whole modeling discussion is what do we believe about our drug? So, you’ll have a, often a target product profile about here’s what we think our drug can do for these patients. Here’s what we think the effect is going to be, et cetera. And that’s going to percolate into a, some assumptions about the trials, but also some assumptions about the population on the back end and the overall revenue opportunity. Also, into some assumptions around pricing. But then I may have some different models, well, if everything goes really well, then maybe I have an upside scenario of how well my drug works. And that looks totally different. And then maybe I have a downside, a more conservative scenario where the drug works but not quite as well as we had hoped. And then another thing that happens again in big companies is maybe I have two options. Maybe I have one way of developing this drug which would be looking at the mild cognitive impairment population first and then going after other populations later. Maybe I have another way which is the reverse, looking at the more severely afflicted patients first and then if it works there developing the other parts as a life cycle management, and I can make those fist fight against each other a little bit in a portfolio review committee. So, there’s an interesting interplay, separate from how investors look at this, which is a little bit more just sort of analyzing the dollars, in a company this tends to be part of an overall discussion about not just how much it’s worth, but how are we getting there.
Mel Whittington: Interesting, that how I would say does not come into play in the cost effectiveness analysis, even population size. So, cost effectiveness analyses typically don’t even factor in population size. Now, budget impact models, that’s where we can start thinking about population size and who’s the eligible population, because that’s really getting at magnitude of the budget impact. But, cost effectiveness analyses are typically all removed from population size and there have been some suggestions of maybe a higher threshold for really small population size, but population size is usually not a key factor, which I found so surprising of being a big difference between the models that are used in the financial world where market size and population size is so integral to it not really coming into play in pharmacoeconomics.
Frank S. David, MD, PhD: Yeah, I mean, what I find interesting is that even though from what I’ve seen, cost effectiveness groups are not necessarily brought in to give super rigorous and detailed input in a large company at very early stages of development. My sense is that there’s more and more a desire to bring them in to at least weigh in on some big aspects of, of how the cost effectiveness could be affected by some of these different scenarios, right? So, when you’re talking about one of these cases like what we just talked about where you’re talking about, different options in different patient populations where you might be expecting or modeling different types of effect sizes. It can be super useful then for a cost effectiveness group within a large company to kind of in conjunction with commercial colleagues start to figure out, okay, like what’s plausible for pricing in, if we think that the drug ends up looking like A versus B. So, I think that’s more and more starting to get incorporated early. I think there was a, when I first started here, and again, this is not really my core, this is not where I spend most of my time, but historically, I think that a lot of the groups that would have done internal cost effectiveness type of work, were getting engaged very late in the R&D process at big companies. And what I hear is that those groups are starting to be pulled in earlier because there’s more of a sense, especially when you look at areas like Europe, for example, where cost effectiveness is going to be integral to whether you actually get on the market or not at the price that you want, just doing a little bit of a sense check and making sure that whatever the development team is proposing to do is actually kind of plausible from a cost effectiveness point of view on the back end.
Mel Whittington: Yeah, that’s great to hear. And I think it does relate to kind of the timing of the two fields. I feel like with biotech investment and development, there’s a lot of this modeling that’s happening very early. We’ve kind of talked about this already that this can happen at phase one, even before that, whereas, certainly cost effectiveness analysis, there can be early phase cost effectiveness analyses that could be done there as well. But it’s far more common to see a cost effectiveness analysis kind of a post phase three or right around the time of approval or shortly after the time of approval and so I think that is where, maybe how this timing impacts some of this separation between our two fields is partially related to the timing. And so, I’m encouraged to hear some of that is brought in earlier.
Frank S. David, MD, PhD: Yeah, I mean, it’s just all about what questions in a big company that you need answered in order to drive resource allocation, right? So, there are questions that are really important in phase one, to figure out, are we funding this phase one trial or a different phase one trial? And that’s, it really is a cage match in a large company because there’s a fixed amount of resources and there are usually more opportunities than there are dollars to fund them. So, some things are going to fall below the line. There’s some amount of kind of forced ranking that’s happening. Whereas later on, when you’re in phase three, sure, you need to turn over the cards and see whether your phase three works, but it’s not so much about decision making anymore. It’s more about tactical execution, right? You’re not sort of thinking, “oh, our phase three work. Now let’s sort of do the math and figure out whether we’re going to submit this for approval.” Uh, uh, you’re going to submit it for approval. You know that the goal is to turn on the commercial spigot and make all of this happen. So, it’s a little bit more about tactical execution. And then it’s about again shifting the focus more to the CFO as opposed to the head of R&D, where it’s about how much are we actually going to promise to the business that we can deliver from this? And how do we maximize the amount that we can deliver to the business?
Mel Whittington: Fascinating. can we talk a little bit about market dynamics? And that is how, in your models, how do you, or do you not consider uptake and future competitors and price changes. I know we’ve talked about this a little bit, but I would like to spend some time talking a little bit formally about that.
Frank S. David, MD, PhD: Yeah, I would say most project teams, especially in big companies, and I think investors do this too, are constantly updating their models as information comes out, right? And that information is both around your product and around everything else in the environment. So, these are not static things. It’s not a one and done. So, I say that that manifests in a few ways. So, it’s simplest form, if you’re in phase one and there are a bunch of other assets, maybe in phase one, phase two, phase three, you have some assumptions about what that future treatment landscape is going to look like and how that’s percolating down into market share, for example, and then if some of those things unexpectedly fail or unexpectedly succeed, then that might change a lot of your assumptions around what the size of the prize is on the back end. It might also change some of your assumptions around your probability of success. If you’re pursuing a mechanism of action that’s similar to a drug that’s ahead of you and the drug that’s ahead of you succeeds, then maybe that gives you more confidence in your drug succeeding and therefore you’re going to adjust your model appropriately. The other place where I’ve seen sort of the market dynamic, where I’ve, the other place where I’ve heard that the market dynamics are also starting to play in that is related to the IRA specifically, is around drug classes and clinical areas where particular on market drugs are being subject to price negotiation because that essentially is like an unexpectedly early genericization for lack of a better term. So now the whole market has changed in terms of what you think is going to go on at the time that you launch. You might have thought that you were launching in a situation where everybody was branded and sort of competing at premium prices. And now you’re launching into a world where maybe one of the big drugs is a lot cheaper, and that now changes what your expectations are around how much of the market is now, how much additional market share is going to be captured by that drug that maybe they wouldn’t have been, they wouldn’t have captured otherwise, and how is that affecting what you’re going to be left with. So, I think that there is a constant reevaluation of these models by both investors and by the manufacturers during the whole R&D process to try to continually take both the progress of your own drug and also the external events into account.
Mel Whittington: Well, the reevaluation in the world of pharmacoeconomics and cost effectiveness analyses, I think reevaluation isn’t all that common, surprisingly, that it’s kind of, okay, done oftentimes at the time around launch. That’s when, you know, and, you know, price might be negotiated and then certainly can be revisited. And I think a lot of times it should be revisited with new evidence or with more certainty, but reevaluation, I would say, is not as common in the world of cost effectiveness as it is and how you just described it.
Frank S. David, MD, PhD: Yeah, I’ve always found that a little bit weird, actually, because a lot is happening later, and you’d imagine the cost effectiveness of a drug actually does change. So, for example, when you make more data, you maybe know more about, you have more certainty around what the real clinical value and therefore economic value is of the drug and somehow that should play in and then similarly, as you said, if you’re thinking about sort of the out years when a drug goes generic, then if you change your model in terms of when you think that’s going to happen, you change your assumptions around when you think that’s going to happen, then certainly that should also change what you think about the value.
Mel Whittington: Well in convectional cost effectiveness, we just keep it static, and we don’t incorporate genericization and so we don’t have to change it.
Frank S. David, MD, PhD: Yeah, I mean, which I guess is fine for that particular moment. Unless you think that the lower price in the out years is really what changes the whole, the whole value of the asset, right?
Mel Whittington: Right. And in a lot of our models, it is over a patient’s lifetime. In a recent CPA exclusive, we looked at the starting population, I think was 48 and they could be on the drug for their lifetime. And so, there’s certainly a situation for even people starting it at launch. There’s a chance that they would be on the generic version should it enter and, 14 years or if even an IRA negotiated price after nine years, something like that and so a big advocate for incorporating genericization into cost effectiveness analyses as it brings it a little bit closer to reality. Again, that’s not going to be the right answer. But does it bring it a little bit closer to reality to help us understand? I think so. And I think it’s something I hope our field will do a better job.
Frank S. David, MD, PhD: Yeah, there’s something that there’s something to be said for just having good face validity of a model, right? That just on inspection, you should say, “yeah, this basically seems to reflect reality, and we don’t seem to be missing anything big.” This comes up actually a lot in early biotech models where part of the discussion ends up being it figuring out which inputs we really care about versus not. So, it’s not that we don’t want it, the classic example is, adherence and compliance. So obviously no drug has 100 percent adherence, right? And 100 percent compliance. Some number of addressable patients who actually get a script either don’t fill it at all or they fill it sort of intermittently. But, in the early stages, I think people agree that we at least just need to nominally think about it so that we can decide if we’re in an area, a clinical area, where this is a really big deal and we need to think about it a lot. Or maybe there’s something about our drug where the whole value proposition is improving compliance, right? Because maybe we’re doing a once monthly injection as opposed to a daily therapy and we think that that’s going to improve compliance. So, in those kinds of cases, the fact that it’s there at least makes you stop and think and decide whether you want to include it or not. And I think similarly for these types of, for the points that you’re mentioning, it’s good to just sort of stop for a minute and think “Do we need to pay attention to genericization and how that really is affecting the overall value?” There may be some cases in which we think the answer is no, but there may be a lot of cases in which we think the answer is yes. And I think it’s always worth just making sure that on its face the model seems to be reflecting what you believe is important.
Mel Whittington: I agree. And then to bring it back to what we were talking about the beginning of the episode about the tornado diagram. Great, let’s include genericization, assign a huge uncertainty around it and the tornado diagram and see, is it in our top 10 as our things that matter and the cost effectiveness analyses. And if it is, then that’s telling us something, if it’s maybe it’s fine that it’s not omitted. But at least in the models I’ve built, it does tend to matter dramatically. So, I think it’s something that will continue to see the pharmacoeconomic field grappling with related to that, I think there’s even kind of unique considerations of, do you incorporate it when you’re thinking about cost effectiveness, but then when you’re thinking about the price to the manufacturers, is that different. And so those are things our field is still just beginning to grapple with in pharmacoeconomics. And it’s something that I’ve always been, ever since I’ve started talking to more investors and people at the investment bank and innovators, how integral considering genericization and accept that eventually, the price will hopefully fall off a cliff and competition will enter. How core that is to your modeling and how core it is not to current modeling in pharmacoeconomics.
Frank S. David, MD, PhD: Totally.
Mel Whittington: All right. Well, we are almost at time. I want to close with, you know, acknowledging that clearly I’m biased toward pharmacoeconomics, but I think, pharmacoeconomics and healthcare investing and development are both very important to the healthcare industry. You and I both talked about the importance of resource allocation and resources are finite, whether that’s our pooled healthcare resources via premiums and taxes or, resources of investors and companies and thinking about what to, what to develop, and those resources are finite, and we need to think about how we can efficiently use them. And pharmacoeconomics can provide interesting evidence to inform efficiency conversations and value and impact. And obviously the work that healthcare investors and developers do and the expertise and resources they provide in bringing new innovations to market is essential. Our fields, they have been talking past each other, I hope that stops. I think our worlds are beginning to collide a little bit and I think largely this is because of the Inflation Reduction Act. There is now in the U. S. government negotiated pricing in the patent protection period and that changes things. the kind of rules of the game have been changed and unfortunately there’s still a lot of uncertainty around that, but I think, hopefully we’ll continue to learn more over time. So, as our worlds are beginning to, to converge between pharmacoeconomics and healthcare investment and development, I’m so glad to know you, Frank. I’m glad you were willing to come on the podcast and I think really through collaboration and shared understanding between our two worlds is the way to get us to an efficient healthcare system that still incentivizes healthcare innovation.
Frank S. David, MD, PhD: Yeah. Thanks for having me. This has been super fun.
Mel Whittington: I do have one more question, if that’s okay. We always like to end our episodes with the following question, and that is, what is the best piece of advice you have ever received?
Frank S. David, MD, PhD: Yes, and thank goodness you prepped me on this because honestly, uh, I think I would have been paralyzed. But I did give it some thought, and it didn’t take that much thought because it’s something I have taped up to every computer monitor that I work at. My friend Alice Pomponio, who now runs the venture arm of American Cancer Society called Bright Edge, was a colleague of mine at AstraZeneca. And she recognized that for me, one of my big problems is I am very easily excited and distracted by many things and okay, so then maybe this will be good advice for you too. She sat me down at lunch one day and she said, here’s what you need to do. Few, Focus, Finish.
Mel Whittington: Okay.
Frank S. David, MD, PhD: So, I have that taped up to remind me. It does not work very well. This is still a growth area for me. Let’s just say, but, I do have it taped up in all of my offices to remind me that I need to you get things done as opposed to starting new and shiny and a little bit more attractive things because as you know the last five or ten percent of anything is usually really hard and somewhat horrible and not fun and it’s always more exciting to start something new. But I’m trying to fight against that, so that advice has been super helpful to me in the to the extent that it’s worked at all it’s been super helpful to me.
Mel Whittington: That’s great. I have your guidebook right here on my desk, so I’m going to write it on that, and it can be my reminder for multiple things.
Frank S. David, MD, PhD: Excellent.
Mel Whittington: Well, thank you so much. I appreciate you being willing to do this, and I hope to talk to you soon.
Frank S. David, MD, PhD: Excellent. Thank you so much.
Thank you for listening to this episode of Perspectives. If you’re interested in participating in future podcasts or would like to learn more about the Leerink Center for Pharmacoeconomics, please email cpe@medacorp.com.
