 Well, thank you everyone for your promptness in returning to see it, so we can go with our morning closing panel, and I have the great pleasure of moderating this one on paying for value, which we initiated the discussion in the previous one. We're pleased to have a very diverse set of perspectives represented as we take an honest look at how to finance and pay for drugs and including what it will take to make payment mechanisms work in the real world. I'm going to introduce the panelists on mass, and they will then just present in turn, and then we'll have a bit of a moderated discussion, and we'll hopefully have a good amount of time for questions from the audience. Let me start immediately. To my left is Amit Serpatari. He's an assistant professor of medicine at Harvard Medical School, an associate epidemiologist at Brigham and Women's Hospital, and assistant director of the program on regulation therapeutics and law, Portal, and probably took you a while to come up with that one, within the division of pharmacoepidemiology and pharmacoeconomics. His research draws upon his interdisciplinary training, my God, you guys are testing me on these words, as an epidemiologist and a lawyer and focuses on the effects of laws and regulations on therapeutic development, approval, use, and related public health outcomes. To his left is Melissa Abbott. She's a clinical pharmacist for pharmacy management consultants, a division of the University of Oklahoma College of Pharmacy, where she is a key decision maker with the design and implementation of an alternate payment model's value-based contracts for Oklahoma Medicaid. Dr. Abbott is also a primary presenter of clinical reviews and criteria reports for presentation to the Drug Utilization Review Board and Oklahoma Health Care Authority. To her left, Anand Kapoor, is the executive vice president of Market Access at MKO Global Partners. As a partner and co-founder of MKO Global Partners, he leads pricing and market access projects domestically and globally. Anand has focused his career on commercial strategy and analysis in the biopharmaceutical industry, including commercial launch planning, pricing, market access, and life cycle management. And then batting cleanup is Jennifer Day. She's with Kaiser Permanente's National Drug Information Services and serves as coordinator of the Emerging Therapeutics Strategy program for KP National Pharmacy. Dr. Day's areas of expertise include drug information and analysis of clinical trial evidence, development of evidence-based guidelines, strategic planning for the use of biologics and emerging pharmaceuticals, biosimilars, especially pharmacy oversight, and pharmaceutical pipeline forecasting. So with no further ado, I will let Amit come up and take us away. Hello. Oh, by my own. Ah, great. Well, good morning, everyone. It's a pleasure to be here, and thank you to Kaiser Permanente for the chance to talk about this issue. In terms of paying for value, I think I just want to get started a little bit with some information we've seen before, but just to put this in context, what is the impetus behind wanting to go to value-based drug pricing? And here in the U.S., when we talk about rising net U.S. prescription drug spending, in 2016, if you put together retail and non-retail spending, as Mark Miller mentioned, we are approaching close to $500 billion in drug spending. From 2017 to 2026, there's predicted that there would be a 6 percent annual increase in net retail spending, which would be faster than any other major health care good or service. And what's really driving this? Well, drug prices. So the median annual list price of a new cancer medication in 2013 was $87,000 in 2017, and today it is over $160,000. We've all heard about Zolgensema and the issues in terms of the burgeoning pipeline of new gene therapies that will soon hit the market and what we're going to do with that. But when we put this into context, what are we actually paying for with these high list prices? Well, there's little correlation between price and clinical benefits. So one study I took part in, which just evaluated 138 new drugs that entered the German market between 2012 and 2016, all of these drugs entered the U.S. market as well, that the German AMNOG process, which I'll get into in just a little bit, but found that 60 percent actually ended up with a negative relative benefit assessment. You've heard a little bit about a study that Reena Kanti, Peter Bach, and Howard did that found that even if you account for the clinical benefit of cancer drugs, cancer drug prices are increasing independent of any association with their clinical benefit. So in this context, in the context where we are paying more than any other country per person on drug prices, we need to do something different. And what is different is potentially what we'll call value-based pricing, which we've talked about a little bit in the last session. What are the goals? Well, the goals are more efficient spending and perhaps more targeted signals to innovators about the drugs we actually want to bring to market. And so if we take a little bit of a cue from what the European landscape is here, there are really two approaches to value-based pricing, two generalized approaches to value-based pricing that we see in Europe. One is the English system or the NICE system, where what we're talking about is what the added cost is per benefit gained. And that benefit is a quality, a quality-adjusted life here. And what we're really asking is what is it worth for us to get an added year of perfect health? And there's various ethical dialogue that can go into that. But we can say that between $20,000 and $30,000 per a quality-adjusted life here, which is what NICE uses as a loose threshold and will evaluate the drug before it enters the market. And that there is no arbitration process for a disagreement there. But then there's an alternative system which says let's not focus on comparative cost effectiveness. Let's just first ask ourselves, relative to what exists on the market, is this a major added benefit, or is it a minor added benefit, or is it even less of a benefit than what exists on the market? And depending on what that benefit is, independent of price, then we'll enter a potential negotiation on the price. And for those that have no benefit, we're going to do reference-based pricing. And here in the German system, what they say is, all right, let's not have this interfere with drugs getting on the market. Let's let the drug get on the market, and let's do this assessment while in the first year while this drug is on the market, and then this will apply subsequently. So when we're talking about what evidence we need, what are the administrative costs and time associated with making these types of decisions, these are just two generalized approaches to handle the issue of value-based pricing. So in the U.S., what are some examples we've seen? Well, and I think I'm going to, I'm going to right now go ahead and say outcomes-based pricing is an example of value-based pricing, though I don't personally think it necessarily is, and I think Peter Bach has mentioned that as well. But these are a scheme of what people will talk about when they usually say value-based drug pricing. This is comparative cost effectiveness. Similar to the NICE model, you could take a look here, Regeneron and Sanofi, utilize this type of system approach to ICER to say, well, what is the relative, what is the comparative cost, what would be a fair price for this drug based on its improvement in quality-adjusted life years gained? And what that sort of negotiation approach yielded was it ensured that there wouldn't be strict utilization management on that drug. And if you take a look at the sales of the drug in 2018, you had almost a billion dollars in sales. You could have indication-specific pricing where you actually say what is the value of this drug for the specific indication it's being approved for with the knowledge that specifically for cancer drugs, we know that in certain indications the bang for the buck is going to be greater than for other indications. And so here is an example of express scripts doing an experiment with oral anti-cancer drugs where they would calculate a weighted average based on estimates of indication-specific use. And finally, you have outcomes-based pricing where a classic example was with Repatha, where you would say, look, if this drug actually proved to work after a certain period of time or not work after a certain period of time here, it was that the company would refund payments for patients who have heart attacks or strokes after at least six months of on the therapy. So this is just a broad landscape of the types of examples of value-based pricing that exists. And this is what I really want to focus my time on in the last three minutes is the logistical challenges here. So one of the things that you'll probably hear is that Medicaid best price rule is a challenge for value-based pricing. And I just want to put up one reference which sort of pushes back on this notion, well, what is the Medicaid best price rule? Well, it says that there must be, you're going to give Medicaid the best price. And so that is going to be the greater of in terms of savings, 23.1 percent rebate off the list price, or the best price on the market, which averse greater. So there are some concerns in the case of indication-specific pricing that the best price for one indication would then apply to all of the other indications. That's one concern. Another concern is for outcome-space pricing, for a specific drug and a specific patient that didn't work, if you refunded the price of that drug, the price would be zero. Does that mean Medicaid effectively pays zero for that drug? Well, no. The argument in this legal analysis is that there are ways to get around this under existing rules. And so the need to actually do away with best price is not really there. So you could use a weighted average pricing, which would take care of indication-specific pricing issues concerns. And you could calculate, or in your contract, say that the rebate is based on performance across the total population. I can get into more detail about those specific issues. But the other challenge is one that has been brought up quite a lot today, is the limited evidence base for many new drugs, and particularly when we're talking about gene therapy cures, which may not turn out to be cures five or 10 years down the line, in which other payers may be on the hook for those particular patients then. And so of 68 cancer indications approved by the EMA between 2009 and 2013, only 35 had shown improvement in overall survival or quality of life. If we take a look at the certain accelerated approval pathways, we know that drugs are increasingly being approved on the basis of surrogate outcomes. These surrogate outcomes are not actually ending up in certain cases to be correlated with the actual clinical outcome that we want. So what are we going to do here in these instances where we have limited evidence and where the evidence actually shows up later that the drug isn't effective? So how do we, that's one concern. Another concern is the limited infrastructure to track relevant outcomes. So how do we actually know this? Usually we use claims data, but are LDL levels going to be in claims data? They're not going to be. So how do we make sure that we have the infrastructure necessary to track these outcomes? Then there are theoretical considerations. And I think Len Nichols spoke very eloquently about, well, let's back up. Why are we paying for drugs on the basis for value when we don't do this for other areas of the healthcare sector? What is unique about drugs compared to other healthcare goods and resources? And we could argue that because they're intense capital-driven goods that maybe there is something that warrants value-based pricing. Is it okay if I take one more minute? Okay, thank you. So then what is the appropriateness, and we talked about this as well, of allowing manufacturers to extract the entirety of the value? One point that we haven't talked about, and this is particularly salient for these new gene therapies that are coming on the market, is the amount of taxpayer support that has gone into their development. So about 25% of small-molecular drugs approved between 2008 and 2017 were based in part on patents or other late-stage contributions, and I emphasize late-stage contributions where I'm not just talking about the preclinical work that is done at NIH. I'm talking about studies in humans that are done by NIH or by public-sector research institutions. So about 25% of small-molecular drugs have some of this late-stage contribution. And why is it that we are extracting the full value and allowing that value to go to manufacturers in terms of how we structure this whole system? So then I want to focus specifically on theoretical considerations for comparative effectiveness, indication-specific pricing, and outcome-space pricing really quickly. So in comparative effectiveness pricing, we've got a general problem if the comparator offers limited benefits and is highly priced. In indication-specific pricing, and I'm sure Peter can weigh in on this, there's a retort that indication-based pricing results in higher prices for patients who benefit the most, higher utilization by patients who benefit the least, higher overall spending, and higher manufacturer profits. That may be true, but I think that that critique also negates the fact that indication-specific pricing, if it is based within the context of a reformation of the system that doesn't assume monopolist pricing but assumes that we are fundamentally pricing based on value and then indication-specific pricing, that would be a little bit different. But this is a pertinent critique that we need to deal with. And then with outcomes-based pricing, there's the possibility of illusory savings, which was hinted at, but I want to get into detail a little bit more, which is if the initial price isn't actually based on value, it doesn't matter if you get a full refund if it doesn't work. On net, you're still paying a ridiculous amount. And depending on how that outcome is structured, if it's only assessed six months later, whereas the real clinical benefit would be after one year or two years, then that structure of that contract really isn't driving value in terms of what we need in the system. And I think I'll stop there, but hopefully we can engage in some of these issues in terms of some of these challenges. That's not to say that there are great opportunities here. But what it is to say is I think the comment that was made earlier that we really do need to step back and ask ourselves first, what are the principles we want for value-based pricing? What do we want it to do? And then are these specific ideas that we have for subsets of value-based pricing actually accomplishing those objectives? And if not, is it time to drop some of them? Or are there ways in which we can reform those approaches to actually get us to where we want to go? Thank you. Good morning, everyone. So I'm Melissa Abbott. As mentioned, I'm a clinical pharmacist with Pharmacy Management Consultants. Our role at PMC is to support the Oklahoma Health Care Authority and their administration of pharmacy benefits for members of Oklahoma Medicaid. So the reason that I'm here today is because Oklahoma Medicaid was the first state to enter into an outcomes-based payment arrangement for pharmaceuticals with a drug company. So today I'm just going to touch briefly on our experience with entering into these some of the challenges that we've encountered along the way. But before I dive into our individual experience, just wanted to touch on really the focus of alternative payment models, value-based contracts in Medicaid. So as we all know, prescription drug spending is continuing to rise. We're really at a breaking point. We're at a point where this is going to be unsustainable from a payment perspective, especially for payers like Medicaid, where we have a finite budget. We're really trying to stretch that dollar as far as we can so we can treat the most patients possible. And as we all know with Medicaid, we're dealing with a very vulnerable population. So we want to make sure they're receiving the best valuable health care that's available. So this just shows a trend in the number of novel drugs which have been approved. But as you can see here, during state fiscal year 2018, Oklahoma Medicaid spent over 40% of our total pharmacy expenditures on less than 1% of claims for medications which cost over $1,000. So next I just wanted to touch on some of the barriers that we've encountered along the way. We've encountered quite a few. But the main thing I want to highlight here is just the resources that have gone in to making these contracts work and getting them in place. So these value-based contracts, as I'm sure all of you are aware and starting to realize, they're very complex. And they are a lot more complex than they do seem on the surface. So as you can imagine, simply defining the outcomes that are sufficient indicators of efficacy for these drugs is a challenge. But then when we add the level of assigning a rebate to those outcomes, that just increases the complexity and the difficulty in arranging a negotiation or a contract. But this is further complicated and it was touched on just a moment ago. But what data is accessible in the first place? So from a payer perspective, we're really looking for something that we can define, that we can measure. And what we can measure is in our claims data. And right now we're limited to our pharmacy claims data and our medical claims data. So unfortunately, we can't draw in those labs if we're wanting to contract around a diabetic medication. We can't look at Hemoglobin A1C as we don't have access to that level of information currently. So really tied to that healthcare resource utilization via either medical claims or pharmacy claims. So just a little bit about Oklahoma. We serve approximately a million members on average over the year. We are 100% fee for service. So we do not have any managed care organizations currently in this state. I will say this has made the negotiations for these value-based contracts a little easier because we're dealing with one payer, one manufacturer, not needing to draw in several different MCOs in the negotiations. We are also a member of a purchasing pool. So this is important because we do have to keep this in mind as we're negotiating these value-based contracts. How are they going to impact our supplemental rebates that we have through our purchasing pool? So something that we always are keeping in mind. And then again, just pharmacy benefits managed by PMC. So with us being a part of the University of Oklahoma College of Pharmacy, I will say that this has been a key to our success in being able to implement these contracts because we do have access to the data side of the university. So our grad students, our PhDs, they're able to help us when we go to analyze the data. And so it's definitely been a benefit for us to have that relationship with the college. So we've had several partners along the way. I've discussed PMCs as well as the health care authority's role in designing and implementing these contracts. But we've had several other partners along the way as well. So we've participated in the State Medicaid Alternative Reimbursement and Purchasing Program, or SMART-D, to encourage the use of APMs in Medicaid. And then the National Academy for State Health Policy also awarded a grant to Oklahoma in order to develop a value-based purchasing arrangement. And then finally, we've partnered closely with drug manufacturers to work through these contracts. We've discussed these proposals with several manufacturers to arrive at the ones that we have in place currently. So just a little bit about our approach. So when it was decided that we wanted to go down the road of trying to implement a value-based or an outcome-based payment arrangement, our goal was to negotiate a mutually beneficial contract with a manufacturer in hopes that this would pave the way for other Medicaid programs and states to also enter into these types of arrangements. So we've definitely tried to share our lessons learned along the way. And I will say that in going through these negotiations with the manufacturers, really anything has been on the table. We've been open for discussion with all therapeutic classes, all drug areas, really trying to look at the best options for our members. And then the timeline here. So we've been working through this process for several years now, but we actually received approval of our state plan amendment from CMS in 2018. And that really gave us the go-ahead to go ahead and establish these value-based arrangements. So I'll discuss a little bit about the different contracts that we currently have in place here. But as you can see, we do have a couple of collaboration agreements outside of the value-based arrangements. So those collaboration agreements are actually population characterizations focusing on multiple disease states. So really, we're taking a deeper dive into the data to see where we may have gaps in care or therapeutic areas that we may want to focus on. And then hopefully, ultimately, arrive at an outcomes-based arrangement as a result of those collaboration agreements. But as you can see, we have a couple of arrangements with different pharmaceutical manufacturers around long-acting injectable and i-psychotics. So initially, we focused on what I would consider a simpler metric for us to work through. And that would be adherence. That's something we can easily get through our pharmacy claims data. It does not require us to look at medical claims. So to get our feet wet with those initial contracts, we did start, like I said, a simpler metric with adherence. But we are continuing to progress those contracts to hopefully look at more of a clinical outcome, tie in those medical claims, potentially looking at things like reduction in hospitalization, or what is another outcome we can look at for an LAI. Another one of our contracts is around injectable antibiotic. So again, on this one, we're looking for overall costs and potential savings. So again, tying in those medical claims, as well as the pharmacy claims, and hoping that our overall spend is decreased. So I will say that I'm cautiously optimistic on the outlook of all of these contracts. The financial benefit in the short term has been modest, but we have learned many lessons along the way. And do hope to continue to progress through and implement additional contracts as we continue to move forward with this. So we'll continue to work on identifying other therapeutic areas, but that's all I have for right now. Thank you. Thank you. Well, thank you, everyone. My name's Anand Kapoor. I'm an executive VP with two labs, pharma services. And the way I think about value is perhaps in a different way. I think of value as a pretty complicated Venn diagram, where there's a bunch of stakeholders occupying a bunch of different circles, and there's some definition of value in between that we're all trying to get at that each stakeholder will extract some kind of value from. At two labs where we sit is in the place of each of these stakeholders tend to be physicians, patients, manufacturers, and payers. And as we think about value, I have to first start by congratulating and applauding Kaiser for actually bringing this group of stakeholders in the room together. I think that probably needs to happen on a national level. And these are messy topics. They're very, very difficult to answer. We can look at what I'll talk about today is how some other countries have attacked this and what they've done to define value and the specific ways they've done that. But I would argue that we are actually in a period of transition. That we're actually moving from a very simple concepts around defining value from a pharmaceutical perspective and from a payer perspective around adherence models that we just talked about, around disease level initiatives, around diabetes, for example, around cost savings initiatives. And we are in this moment of transition where things are becoming much more complex. We're talking about risk and outcomes-based agreements. We can talk about whether that actually defines value or not. And we can talk about other elements like the supporting infrastructure that is continuing to evolve, the data that is required to actually execute these types of complex agreements. I think so far it's been super interesting because Zuljansma has been brought up as an example. And even that as an example, we are coming at two very different definitions of value, almost at opposite sides of the spectrum. So where is the overlap there? Because there must be some kind of overlap. And I think part of the challenge is it's actually we need to get a sense of how difficult the definition of value is. Because to me, you actually have to get quite specific. It's very difficult to talk about value out of general sense. You have to talk about an indication, or a product, or specific outcomes that you're looking for. Because if you don't do that, we can talk about patterns of where we're trying to go, but it gets very detailed very quickly. Typically, the pharmaceutical definition of value has been v equals r plus or minus d. v is the value of the current therapy. r is the reference drug or the current standard of care. And plus or minus d is the level of differentiation it brings on the current standard of care or that reference therapy. So that has historically been a generalization. It may or may not apply in all cases, but that's the challenge of how pharma has defined that. Typically, they look at cost offsets. Let's look at Hep C, for example. When the Hep C manufacturers were talking about how they're going to price some of these therapies, they came up to let's talk about the value of decompensation if patients aren't treated. Let's talk about the value of liver transplant, way down the line. How do we assess that value in our current pricing? And it was discussed before that the level of competition despite the level of competition, the cost can go up. In some cases, based on this transition, we've seen costs go down in Hep C, for example. They started off at 60,000 to 70,000. They've increased for a period to 120,000 to 130,000 when two branded agents were used. And now they've dropped to somewhere between 20,000 to 30,000. And so the point being is there are examples. And we are in this transition phase where we're moving towards some kind of definition and we're getting there. I think the difficulty is within the US system is the number of circles. Is we have GPOs and specialty pharmacies and specialty distributors and compendia organizations. The number of stakeholders within the US is unparalleled with the rest of the world that, although it's a bit of a misnomer to call them single payer systems, they're far less stakeholders in Europe and Canada and Australia, for example. So some of these become a bit more easy to execute on. And why are we even talking about this now? I mean, these drug prices are growing. And this is just a subset of them. If you look at the full specialty pharmacy pipeline, it seems that every product is increasing price, is going after a rare orphan disease. And the number of these agents is growing significantly year on year. And so as you compound this, the payers, for example, look at this and they get very, very nervous. How are we going to pay for this? How are we going to come up with different options to pay for all of these different agents? And what does that going to mean for our PMPM numbers and our overall cost of patients? We've talked a little bit about some of the innovative payment models. We've talked at length at some of these. I think one of them that we can talk about is this one year milestone payment, where Spark and Laxterna came up with this idea that if they fail to meet the endpoint in the short, the 30 to 90 day short term, and they also fail to meet it with long term durability, then there would be a rebate. So we are heading in this direction where we are taking into account the short term and long term considerations, but there are still challenges. If, for example, on these gene therapies, there's a breakthrough, nine or 10 years from now, and that person needs to be retreated. Who bears the cost of that, given that a plan member only lasts somewhere between two and two and a half years on a plan? They've cycled through plans. I think the benefit and the appropriateness of having this conversation at Kaiser is they may not face some of those challenges or face those challenges differently. So even we see within pairs, IDNs versus other commercial pairs, they're differing definitions of value. So we talk about pairs as a group, but there are subsections within that. Let's talk about within the patient segments. An MS patient may not see any value in a first line MS therapy, but they may see that doesn't help them at all from an MS perspective. They may see value in later line therapies, second, third line, where that is the benefit that is being brought to them from a clinical perspective. So we need to think about, even from a patient perspective, where they sit in the treatment algorithm, their definition of value is different. And I think lastly, we've talked a lot about cell and gene therapies. I think the payer thought on this is that what are we not talking about? Why are we not talking about the 30 million diabetes patients or the 30 million cardiovascular patients or the 50 million patients that are currently suffering from other very comorbid indications, and these costs are also going up over time. So although the cell and gene therapies are definitely fodder for discussion in terms of cost, the uptake has been relatively low. Pairs are now paying for these on a one-off basis, because they can for now, Zoologisma, for example. Although there are some contracts in place from a value-based perspective, pairs are looking at these from a case-by-case basis. There's not enough uptake within that product, for example, to actually justify a value-based agreement. And so we need to think about all of that as well. I think recently, within the last few weeks, SIGNA announced a new program called Embark that allows employers and insurers to pay a per-month fee for a service that will cover the cost of gene therapy and management for their use. And so again, we are on this pathway to this transition point in the future where we talk about how to very specifically define value. We've talked a bit about what is working. I'm going to talk about what is not working, the five-year annuity payment, for example. So this is a challenge. I mean, in Europe, it is hitting a challenge because they're not necessarily seeing the uptake. And here, because of the fact that the plan members change every two and a half years on the commercial side, it is difficult for a payer to justify that value given that another payer, for example, may accrue all of those downstream benefits. So we need to be very, very careful and very specific on how to talk about value. On the examples that are outside of the US, this is happening very quickly. There are a number of examples within the cell and gene therapy space that are taking place here. And granted, from a European perspective, a Canadian and Australian perspective, this makes a big difference because of the complexity involved of executing these agreements. We asked once a large managed care payer around HEPC, what types of data are you gathering? And he mentioned we're gathering SVR data, which is great. The primary endpoint in HEPC, easy to gather. So we asked him to look back and see if he could actually comment on SVR rates. And then he looked back in his database and he actually realized the SVR data that they thought they were gathering, they weren't actually gathering. So there is a technical aspect that needs to be developed in order to make sure that we can follow through on all of these value-based agreements. And in addition to the lawyers who tend to manage risk very differently, we need to make sure they are at the table too. Because if they're not, then you can have these very specific conversations around defining value. But then at the end of it, if the lawyers don't agree, they're very difficult to move forward. So appreciate the time. Look forward to the conversation further on the Q&A. Hello, everybody. My name is Jennifer Day, and I'm a pharmacist with Kaiser Permanente Drug Information Services. And the scope of my presentation is a little bit different than my fellow panelists. I was asked to kind of give more insight into the pharmaceutical evidence landscape. The many challenges we're currently facing and how at Kaiser we've chosen to approach some of those challenges. Now, as the Drug Information Service Drug Information Pharmacists, one of my core responsibilities is to evaluate the evidence and to work with our physicians to determine how best to use that therapy for our 12.2 million members. And before we pay for value, we obviously need to determine value. And in order to determine value, we must first understand the benefits and risks of a therapy. So ideally, we'd like to have at least two well-designed, randomized, controlled clinical trials to help us really have a good understanding of the benefits and risks of a therapy and to really help enable informed prescribing. But more and more, this has changed and that previously established expectation is no longer there. In recent years, the evidence base has shifted so that more and more approvals have less evidence. Last year, we had 59 novel drug approvals by the FDA. And 73% of them were approved via one or more of the FDA's expedited review pathways. Now, this in part reflects the shift in drug development to orphan or rare diseases. These therapies often qualify for one or more of the FDA's expedited review pathways. And in fact, last year, more than half of the novel drug approvals had orphan drug designation. So while these therapies offer new treatment options and fulfill an unmet need, data is often lacking to guide clinicians on how best to use these therapies. And granted, you can't always expect two well-designed clinical trials for therapies that treat very small patient populations, but it is becoming more and more challenging as drugs are approved based on one study. Sometimes it's well-designed, sometimes it's not. Sometimes it's only phase one or two study data, interim study results, and then surrogate outcomes that have yet to establish a clinical benefit. So that's when we get to that right side of the slide, where there's lots more questions than answers when these products become commercially available. And unfortunately, we've been spending a little bit more time on that side of the slide lately than we like. So to talk more about those challenges is the type and amount of evidence available. I mentioned the expedited drug approvals. So often the rigor of the evidence behind those approvals is much less than what would be normally expected or hoped for. Accelerated approvals. Samir earlier mentioned Exxon 51, approved based on a surrogate endpoint, the increase in dystrophin and skeletal muscle, of which we have no idea what it means, how it correlates to a clinical benefit, yet it was approved and they had to labor required to perform a post-market conformatory clinical trial that has yet to happen. Limited or unpublished data. So even if there is data, sometimes it's not published or publicly available. And that actually relates a little bit to the FDA integrator review, which I'll get to in the next bullet point, and I probably should have switched them around. But of course there's unknowns when drugs get approved, right? Standard is long-term efficacy and safety. But sometimes these therapies are being approved without even the optimal dosing or duration of treatment being known. And recently the FDA talked about proposing to move to an integrated review document. So this is basically to streamline and make it more readable. Currently they post or drug approval packages on the FDA website, and unless you review drugs you probably have never accessed it. But if you have, they are super comprehensive documents, hundreds and hundreds of pages of information you don't even know what to do with. But they also include really valuable insight, individual reviewer insight, comments about concerns, how they're rationalized or how those potential safety issues are maybe alleviated with certain types of data, things like that that you don't find in any published article or product labeling. And so some of the stakeholders who use these FDA approval packages have actually expressed concern that this new streamlined integrated review document might inadvertently admit some of these really valuable insights. So we're hoping that if the move is towards this integrated document that those comments still are available to us, especially when there's limited published data available. Now I mentioned that some therapies are approved and have requirements to do the post-market confirmatory studies. And there's been numerous articles published about how these studies are often delayed well beyond the timeline the FDA set for them. Sometimes they're not even completed. And this leaves us with knowledge gaps that need to be filled. We still don't know how best to use these therapies, how safe they are, what value are we truly getting for them? It's been mentioned, I think the earlier panel as well as some of the attendees mentioned broader indications than what was studied. Well, this creates quite an issue, right? Because really we have no benefit risk profile for those patient populations. And our clinicians now feel pressured to experiment in clinical practice. So at Kaiser Permanente, we feel that our integrated medical and pharmaceutical services really puts us in a unique and favorable position. Our national pharmacy works closely with our physicians to provide services in outpatient, inpatient and clinic settings. My department Drug Information Services supports a multitude of informational needs for healthcare providers, members, and our medical care program. I am really fortunate to be working with a team of dedicated drug information pharmacists with expertise in specific therapeutic areas. Together, we follow drugs through the clinical trials through the FDA approval process and we evaluate and assess and review the medical literature and support our tenant of evidence-based medicine. This collaboration between pharmacists and physicians we have at Kaiser fosters a unified approach to the management of drugs to ensure that we use them in the most safe and efficacious manner possible. We also have a fully integrated electronic medical record that facilitates efficient information exchange and decision support communication, as well as access to medical, pharmacy, and laboratory records. Our integrated culture facilitates our goal of accessible, affordable, and high quality care. So what we were observing the recent trends in FDA approvals at Kaiser Permanente, we saw a clear need to be more proactive in evaluating and strategizing for the innovative therapies coming through the pipeline. So we took advantage of our integrated culture and we developed our emerging therapeutic strategy program, which is a collaborative effort between drug information services and Permanente physicians. And together we work to map out practice recommendations and use the strategies to incorporate evidence-based medicine into practice. This diagram walks you through the fundamental steps of our process. First, we identify select therapies to review and close through our drug information pharmacists. We primarily focus on therapies that happen to be very high cost, but also high impact. Maybe they have a very significant safety concern. Oftentimes they have very limited evidence, such as with some of the orphan drugs. And sometimes they may pose significant utilization resource challenges that need proactive planning. And then we identify our physician specialists to work with to review the evidence and map out our plan for optimal use of those therapies within our program. And after formulating and implementing those strategies, we try to close a loop and do a medication use evaluation on these therapies to assess how we're doing. Basically we want to evaluate and look for best practices and optimize patient outcomes. By combining the clinical trial evidence with our internal real data, we hope to inform our prescribers further and basically find those best practices, improve our process, and provide that safe, high quality care that we're all striving for. So I wanted to close with some thoughts on what helps us to be successful. And Kaiser, I think some of the things that have really helped us to be successful in our approach is maintaining our physician pharmacist partnership and promoting our tenet of evidence-based medicine, as well as our physicians resolve to avoid experimenting in clinical practice. Now there are some regulatory areas that could maybe potentially have some improvements, such as tightening the standards for those expedited reviews. I think the comment earlier was incentivizing for evidence development, so that there's less uncertainty. How can those expedited review pathways really support evidence-based medicine? Judicist use of real-world evidence in the drug approval process. Finding the value in the real-world evidence but not so that it complements but not replaces adequate data from adequately designed clinical trials. And finding ways to increase oversight of those post-market clinical trials. How can we incentivize or enforce sponsors to complete those studies so those knowledge gaps are finally filled? And finally to deter abuse of the Orphan Drug Act. Sure, you've all read news articles about companies who've taken advantage of loopholes in the Orphan Drug Act and gotten incentives afforded the Orphan drugs and profited it in ways that were not consistent with the intent of the Orphan Drug Act. So we recognize that these regulatory changes will take time to happen. We look forward to making that shift and moving in that direction so that those changes can help support our clinical approach and the success principles that guide our work at Kaiser Permanente. Thank you, Jennifer, and thank you to the panelists. We have about 15 minutes here and I'm gonna pitch a couple of questions to the panelists directly, but I'll also encourage people who are thinking of questions to consider lining up at the microphone and we'll flip over to you. I have to say if Lead Knuckles is a simple country health economist, I'm a simple city health economist and this stuff boggles my mind with some of the things we're coming up with. I just wanted to make one observation because it's an example of sort of an externality that's present in all of this I think you mentioned the Hep C drugs and avoided liver transplants and that is true for the transaction between the payer and the particular population that's being treated, but as a system matter, there is no decline in liver transplants because there's a shortage of transplantable livers and I think this generalizes to any avoided services that there's a cue for, if you will, so let's think mental health is one example of that. Any of the downstream treatment that you avoid for a particular population immediately gets backfilled by people who are not otherwise getting served and that's not an issue for these particular contracts but it is an issue for public policy to be thinking about, are we really, I mean, that's not a bad thing, right? I mean, it means one patient's being cured and another one is being treated but it does mean that socially we may not see the savings that we otherwise expect. Melissa, I had a quick question for you and it may be my reading of your slide but I think you said that maybe not the primary motive but a motive for entering into these contracts was really sort of a proof of concept. We want to do a couple and get them under our belt so we can do more, am I getting that right? So that was why we landed on the initial contracts that we have in place but the actual thought behind why we want to enter in these contracts at all is just because we are needing innovative ways to pay for drugs, we're at a point where we're dealing with a finite budget, we only have so many dollars to go around, we really need to make sure we're using our dollars appropriately so we can treat the most patients possible. So that's the reason for the contracts but the ones that we have in place, we were looking for a simple metric. These are very challenging. We, entering into these and going into these negotiations, they were a lot more complex than what we had initially anticipated and so just really trying to get a couple so we can get through the regulatory issues, the legal issues, I think someone had mentioned making sure that the lawyers are on board so to get those initial contracts in place but really with the ultimate goal that we do want to move forward with a more clinical or robust outcome so. You never get the lawyers on board and just get them out of the way kind of on top of ours. Well but on that point and this is not just for you specifically but there is a real issue here of the costs involved in structuring these contracts and negotiating them and then the ongoing costs of let's pretend for the moment that there are data available but they will be imperfect data of monitoring and then enforcing the contract. How do you guys think about that? I don't know if you want to go first Melissa or do you want to go to that one? Sure so I think the point is well taken that there are considerable costs with these they also serve a public good in the sense that exploration and experimentation of these then can allow implementation on a more scalable level with other payers and so in my sense it is something that public investment should actually support experimentation of more value based payment experiments and that in general we need better data infrastructure to harness the data that exists but we aren't synchronizing and capturing in an effective way and so for that I think that there are legitimate fears of people wanting to be the first to have to do this and I think that we need to support those efforts and that's where taxpayer money can actually be helpful. And I would agree with that we were fortunate to have the grant from NASSB when we initially entered into these contracts so that helped support get some of that data infrastructure in place. Again being tied with the university has also been a huge benefit for us having some of that data analytics in house and then the way that our relationship with pharmacy management consultants and the healthcare authority has worked has definitely been in our favor but that is always something that we have to keep in mind as we negotiate these contracts is how much, how many resources are we going to have to utilize to not only negotiate but also to implement the contract so definitely a key issue. And maybe just to build on that it's I think the interoperability of whether we call it EHRs or gathering these data sets is probably one of the biggest challenges even in countries where you have a single source of information there's provinces in Canada that have tried this, British Columbia for example that has tried to create a province wide data set for all of its patients and it's been a real challenge even within that small context within a country and a country. And so I think what the implication of that is is a lot of these value discussions can sound great on conceptually and sound great on a slide, they look great. We've ironed out all the details but it's the tactical implementation that can actually prevent all of this and it really boils down to is the infrastructure keeping up with the data requests to execute against these agreements and I think that's a big challenge. I mean Jennifer to that point one of the things that we're sort of blessed with is we have some scale to operate with and we have a pretty data rich environment but I think it'd be fair to say we still encounter a lot of challenges. Absolutely and to add on to what you're saying depending on the outcomes you're looking for and how it's documented it doesn't always lend itself to an electronic medical record. If it's not consistently documented you can't really retrieve that data you'd have to do chart reviews so the return on investment may not be there. So on the data issue and a couple of people have touched on the non-completion or the non-rapid completion of post-market surveillance that FDA has required as part of approval. I mean that's sort of a piece of it to continue studying there but there's still a burden of somebody has to collect the data to be able to enforce these contracts and does all of that fall on the payer side? Should some of it fall on the manufacturer side? Is that something that gets negotiated? Thoughts? Sure so I mean we're on a, as a societal level, we have seemingly come to an agreement not all of us agree but that there is a pressure on the FDA to bring drugs to market faster and I think that if in exchange for that there needs to be as we talked about earlier in the discussion an incentive for manufacturers to collect that data in the post-market and if FDA is not going to implement the requirement or enforce the requirement in a strict way that's one concern. Another concern is pragmatically the concern that why would some patients want to enter into a trial in the post-market space when they wouldn't have to? So there's a logistical concern in general and then there's the concern on price. What is the impetus for the manufacturer if there's not a price war that they might get for collecting the data and it shows a benefit but that comes with a risk too. They have to be willing to be comfortable with the fact that if the data do not show a benefit that that might result in a price decrease. But I think there's ways in which we can structure that system a little bit better to create incentives for better data collection and capture but in terms of who bears that responsibility I think that it's unfair to put that allocation on any individual stakeholder in the system so there has to be some consensus and some sharing of responsibility between payers, between manufacturers, between patients even if they want early access to these drugs to gather that evidence. And I'm not saying there are any easy answers there but I'll leave it there. Are there? One of the solutions that has been put out there is there a third party aggregator? Is there somebody who is impartial I guess if that's the appropriate word to use in this context because the fundamental question is once you gather the data, who owns it? Who's paid for it? Who's gonna have the ability to dissect it and do meta-analyses and fundamentally go in there and actually make conclusions off of it? And there that is a particular challenge. I think there are some solutions being put out there where potentially it's a specialty pharmacy or another group of individuals that interacts directly with the patients. They ensure the patients are being adherent and that's an opportunity to gather data. I'm not suggesting that's the perfect solution but it is a solution of one stakeholder within the system that has the opportunity to in a very clinical setting for some of these high-cost specialty products that they have the patient interaction necessary and they're wherewithal to actually gather that data in a confident environment. And that is a solution but it's by no means a perfect one. I wanna come back to this issue of scalability because I guess as a moderator I'm not supposed to have an opinion but I do. I think A, scaling is hard all by itself but two I think early on, so this is a statement in the form of a question in the form of a statement, early on I'm not sure you wanna standardize and scale. I think we're still in such early days. Maybe you want to be figuring out different ways of data ownership. Maybe you wanna be looking at different cut points in contracts and how do you measure success for different kinds of things? Unlikely that the first couple of contracts out the door are going to be the right contracts. They will need to evolve. So I don't know if you wanna comment or offer any thoughts on what's the length of time that we need to evolve in a space assuming it evolves. So the point, is there anybody else for my answer? Okay, I concur with you which I think why experimentation is necessary. In terms of the length of time that experimentation will take, I think it's an empirical question. It's when we actually see something that looks like it functions well. And so I don't have an answer in terms of a specific number of years, but I think what is important is to allow the flexibility for these types of arrangements to be attempted. And that may mean public support of more experimentation. And I would agree with that. And that has really been our approach is to experiment in different areas, really being open to all opportunities with different therapeutic classes and just really try to see what fits and what's going to work in the long run. Okay. We have a question in the back. If you could just name an affiliation because you're in the dark for the spin. Robert Popovian from Pfizer, how are you? Hi, Bob. So the question I have, and we've been talking quite a bit in the last two panels about capturing value, which is fine. I mean, these value-based contracting mechanisms are fantastic, biosimilars, everything. But I wanna step back and say one thing, and Peter alluded to it in one of his slides, which is the pie chart he showed, that there's about $166 billion lost in value in the supply chain in the United States. So if we're doing all of this stuff to capture the value, what are we doing to ensure that $166 billion is not lost, which is not a small number, right? According to his data, it's close to 40% of the spent, 40% of the growth spent in the United States for drugs. So I wanna hear from the panelists or anybody else who wants to talk about it, like what we're gonna do to be able to get that back into the system. And I'll just add on to that if I may, that that's sort of part of my preliminary question is, are we adding another slice in that pie that has lawyers labeled on it? Well, first of all, lawyers are everywhere in that pie to begin with. So I don't think we have to worry about that. So I wanna just make two, one point, which is that I agree with the slide that was presented that there are problems throughout the pharmaceutical supply chain. I think that there are ways in which we can increase efficiency, decrease spending, actually move incentives more towards products that we as a society need. One thing I am cautious of is that we not get sidetracked in terms of it seems, and again, we can enter hopefully a constructive dialogue about this, but every time that we seem to get into a serious discussion about how to extract value from the pharmaceutical side, from the manufacturing side, there is a lot of arrows pointing over to the PBM side or other areas in the market. And there is no doubt that simultaneous work needs to be done on the pharmaceutical supply chain. But I think we need to start with the most principle fact that what matters most is what the launch price of the drug is going to be. And from there, if that is not tied to any conception of value, we have a problem moving forward and where everything else comes down the line and gets extracted. And so my focus is let's start there. But that's not to say work on other areas of the pharmaceutical supply chain is needed. And I would say that I think that there is actually quite a movement towards reform in those other areas as well. So I'll just leave my thoughts there. Yeah, I mean, maybe building on that, there are, you know, it's absolutely right. I mean, there are players in the system that, for example, they're incentives built into the system that benefit from high cost, high discount type approaches from a pharmaceutical value chain perspective. And I think what we really need to do is start having a bit of a difficult conversation around what value do each of the stakeholders individually bring to delivering very specific value. And I think that we need to couch this argument around, we're not talking about the 14th or 15th ARB. We're not talking about a 505 fixed-dose combination B2. We're talking about a potential hemophilia gene therapy, for example, that could potentially be a curative agent for a bunch of kids with hemophilia. It's very difficult to say that there's no clinical value in that. The price is absolutely debatable. Even in that price discussion, though, it's a challenging discussion. If you take a simple cost offset look at that, and you think about how much a factor in fusion costs for a hemophilia patient over the course of their lifetime, there's huge value in that savings. And if you enumerate that, the value is astronomical. If you take that to a pair, and the manufacturer will say, this will save you money, the pair will say, I can't afford to save any more money. Like you say you're going to save me money, but you're talking about a really long timeline. And so I think we need to really sit down and have an honest conversation around, who are the stakeholders? What incentives are built into each one of these stakeholders with regard to launch price? I think there are better ways to conduct clinical trials, to reduce the cost of clinical trials, and try to reduce that. And I think it's a very challenging discussion, because from my earlier discussion, there's so many stakeholders in this value chain. And if there's going to be some very difficult conversations around who stays and who goes, do you want to follow up? No, I just wanted to say, I really appreciate the first comment, because I think there needs to be a simultaneous discussion on both issues, on the pricing issue, and on the supply chain and the value loss issue. But to say that the price of the medicine is not influenced by the value loss is being naive. I think we should acknowledge that. The reason that some prices are such is because the anticipation is that the value is going to be lost so much more in the supply chain. So I appreciate that first comment that was right on. And that point's well taken. Thanks. And that sort of reminds us of the endogeneity of the price here. And I guess just maybe a quick closing question to you. In any of these negotiations, how do you know you're getting the fair price to use that word today? Put it another way, how do you know that the benefits of trying to hedge on these contracts, on outcomes, and value are not offset by increases in list price? Or is that just a risk or a fact of life that we live with? I'm not throwing that one to you. So maybe can you rephrase it? And maybe I'm not getting it quite. If I'm a seller offering a concession, which is basically I might be insuring against my product, but if I have pricing power, what's against my product's scaling, but if I have pricing power, what's to stop me from raising the list price and then offering? OK, so this is exactly what I was trying to demonstrate was an issue with the illusory savings that might be possible from outcome space pricing. I think it's a different scenario where you talk about what I'm going to call and forgive me. I'm going to call the ICER-based model where you would take a look at the comparative cost effectiveness at the time and base your price on that. But it is true that in all of this, there's the possibility of gaming. And I think this is just generally we see this across the board with orphan drug regulation. We see this with any type of incentive, 340B regulation. So with all of that, there's a risk that somebody is going to win potentially through sneaky ways somebody might win. And we need to develop ways to monitor that system. And so to go back to your question, I think at the end of the day, it is a risk of the process. But it's a risk that's worth taking given the situation we're currently in. OK. Oh, and that's a good reminder that we're not seeking perfection. We're trying to improve on the status quo. So we're time for this panel. So first, I want you to join me in thanking them for their contribution to the day. Thank you. And we're about to break for lunch, but before everybody races for the exits here, we're going to do another poll, if we can, to try and inform our last panel. And the question on the table is, how should we advance thinking and action in the drug value space? And this is not really just a question to help our last panel. It's a question for us to be thinking about as we carry our work forward. And again, you should be able just to text, excuse me, you should just be able to find the thread on your phone if you previously texted this and just enter your answer now. If you weren't here for that exercise, you text KPIHP to 37607. And again, that's texting KPIHP if you don't have the thread on your phone already to 37607. Yeah, if you did it already, for those who can't read the small print on the screen, I'll just read from the bottom up. We need better evidence. No drug works if patients can't afford it. Stop. Maybe amplify with another one. Reducing hospital admissions. Reward manufacturers have stepped up with value-based pricing and payment models. Consolidate the supply chain. Should be a holistic discussion with the entire supply chain. Get all stakeholders at the table. Forms like this help. Now, a specific follow-up. And I'll just leave these up there. And we'll be pulling some of this together for our post-forum materials. So we're going to transition to lunch, which is straight at the back of this room. And we are going to reconvene very sharply at 1.15 for our discussion with Matt Isles and Steve Uble. And given that, I'll just release you to lunch. Thank you, and look forward to reconvening in about just under 45 minutes.