Rae Woods: From Advisory Board, we're bringing you a Radio Advisory. My name is Rachel Woods. You can call me Rae. Today, I want to talk about the life sciences part of the healthcare industry. You may think of these as just the drug makers or just the device companies. But in reality, there is a critical way that life sciences interacts with the rest of the healthcare ecosystem. And that's evidence. So to talk about the evolving role of evidence, I've brought three guests. We've got life sciences experts, Solomon Banjo and Pam Divack. We've also got Lou Brooks, SVP from Optum Life Sciences. Well, welcome to Radio Advisory to all three of you, Pam, Solomon, Lou. It's been a while since we've had a full trifecta on the podcast. Makes it a little bit tricky, but I am excited to speak with all three of you. Solomon Banjo: Pleasure to be here. Lou Brooks: We are excited to be here. Rae Woods: Pam and Solomon, you have both been on the podcast several times, but Lou, you are a new voice for us. Tell us where you come from in the industry. Lou Brooks: Sure. So I come from the Life Sciences Division of Optum, with a background in econometrics and real-world data. And so I lead the Real-World Data and Analytics Organization within Life Sciences Division. And our organization is responsible for curating and licensing all the research-ready data assets that we provide to life science customers today. Rae Woods: Who are life sciences customers? What are the parts of the life sciences industry? Lou Brooks: So from a life sciences perspective, your traditional pharmaceutical manufacturer is the way that most people view life sciences, but it's a little bit broader than that. It incorporates organizations like biotech, medical devices. Funny enough, it even incorporates organizations that work with animals and veterinary-type activities. So all of those particular components, whenever there are activities that are designed to improve people's health, whether that be medicinally or in a device-oriented fashion, those all kind of get roped up into that broader life sciences moniker. Rae Woods: I'm not sure that folks realize that the life sciences industry is responsible for producing the majority of the scientific and clinical evidence that the entire industry actually uses. Pam, prior to the pandemic, how was most of that evidence gathering actually done? Pamela Divack: So in the past, the traditional gold standard of evidence has been the randomized control trial, where you divide up participants into a control group and experimental group, and then you compare the outcomes. For the last several years, there has been a growing interest in real-world evidence, which is any evidence about a product or intervention that comes from outside of clinical trial. So based on data collected from routine clinical practice. Through COVID and other factors, there's definitely been more of an interest and investment in real-world evidence, which has changed the paradigm quite a bit. Rae Woods: Before we get to this new paradigm of real-world evidence and real-world data, which is obviously, Lou's wheelhouse, I do want to talk about that gold standard way of generating evidence, the randomized control trial. Pam and Solomon, you both have been actually calling into question some of the problematic features of the "gold standard." What are some of the challenges with this traditional way that the life sciences industry gathers evidence? Solomon Banjo: So I'd say you can probably think about it in two buckets. One is by virtue of how rigid and controlled they are. They don't reflect typical practice. The patients that they recruit tend to not be as complex as the patients that your average clinician is seeing. And the second bucket would really go around this phase of, what's fit for purpose evidence? And the example I like to bring out is you don't actually need that kind of trial Pam was describing for a parachute. There are other ways we can prove that parachutes work without having that control group. Rae Woods: Yeah. Nobody wants to be in the control group for a parachute. Solomon Banjo: Nobody wants to be the control group. Really hard to recruit for that, which is also a problem with clinical trials in general. And I think when you think about the changing care delivery model, the advancements in data and technology, my real push is we don't always need that. How do we innovate on the model so it's not the sort of static model we've had for decades now? Rae Woods: Because if the evidence system is flawed, it means that downstream decisions about care, downstream decisions about coverage, those are also going to be flawed. Evidence impacts every part of the healthcare ecosystem. My question for you, Lou, is, do you think the rest of the industry has come to this realization that randomized control trials might not actually be the gold standard, always? Lou Brooks: Yeah. I think that the industry and others within the healthcare space have recognized that there are challenges in the way the current model exists. And I'll build on what Solomon was talking about. One of those areas is just in the diversity of those trials and representedness of that. I mean, Solomon talked about complexity of patients. It's a little simpler than that. It's just making sure that the trial accurately represents the makeup of the US population or the population, X-US, if the trial is being conducted there. Organizations are looking to make steps to modify that and change that. But it's going to be small steps. It's going to be small progress to work our way to fixing decades worth of somewhat flawed recruitment practices and policies surrounding those clinical trials. Rae Woods: And fixing the way that we gather evidence doesn't just mean fixing the flawed parts of clinical trials. But it also means expanding to new and different kinds of data collection, which is exactly what real-world data and real-world evidence does. Lou, do you feel like the industry is more willing to accept this kind of data? Lou Brooks: It's growing. The belief and the acceptance, especially post-COVID and being able to see what the industry was able to do, has shifted that forward and gotten people to think more progressively about real-world data as a foundation for evidence generation. And it's not that it hasn't been done previously. It's that it's kind of accelerating that trajectory. I think the challenge still is, much like clinical trials, the real-world data itself has the potential for being somewhat, I'll use the term flawed. That may be too strong a word. But it's missing in many cases, the diversity and a representiveness of the broader population for whatever one very good reason. Lou Brooks: The first piece is it's based on people consuming healthcare. So those individuals that aren't even engaging in the system are totally forgotten from that standpoint. And we need to figure out ways to engage them while we are working with the data that we have today in order to move us to the point where we can generate evidence across the entire spectrum, not just on those that are actually able to engage with the system today. Rae Woods: This is an important point. And I think it's one that's probably difficult for those of us in the healthcare industry who believe in science are going to have a difficult time swallowing, is realizing that the way that we gather evidence, period, across the board, is flawed. And that is okay. It just means that maybe we need both the clinical trial and the real-world data, or we need to continue to test and maybe ask better research questions. And there isn't this kind of perfect Petri dish, so to speak, of scientific answers that we can get in a lab, because they're not going to be applicable to the real world. Solomon Banjo: I agree with the flawed moniker, but I also think it's important to keep in mind the goal of clinical trials traditionally, which has been to get regulatory approval by proving safety and efficacy. And a lot of the work Lou does, a lot of the research Pam has been leading for us, is showing how we need the evidence to inform a lot more than just that. And that's when the limitations really become acute, when it's about delivering on the promise of value-based care, helping address health disparities and other things that weren't the goal of the initial Phase III trial. Pamela Divack: Looking beyond safety and efficacy data, we're also seeing pretty much all parts of the healthcare industry are now looking at things like patient experiences with treatments, or how does a treatment impact total cost of care, or even generate avoidable costs? And we're kind of asking new questions of the data in addition to generating new types of evidence. Lou Brooks: And the question is the most important part. Rae Woods: The research question. Lou Brooks: Yes. It's starting to recognize more importantly, in many cases, people started down the oh-we've-got-to-generate-evidence path without taking a step back and starting with, what's the question that we need to ask and answer? What is the right data that we need to address that particular question? Where do we get it? And then once we've generated the appropriate evidence, where do we take it in order for some type of change to occur? And I think we've started to look more at the questions and really making sure that across the board life science companies, payers, providers, the government, people are uniting around the right question rather than just running down the path of generating evidence. Rae Woods: And that question is about more than just safety and efficacy. So what are some of the questions that you want to see all stakeholders in the industry gather around? Solomon, maybe you mentioned one of them already, which is value-based care. Solomon Banjo: Yes. And I think as we think about that, an embedded question is, what is the differential impact of the treatment across demographics, across different settings? Can we actually say that okay, for Solomon, Pam, Rae, Lou, we have a sense for the actual expected outcome from this treatment to then be saying, "What is a reasonable contract to build around those expected outcomes?" So I think that's certainly one of the bigger questions and where we're starting to see cross industry collaboration at first attempts to say, "Okay, how do we actually answer this? How do we pick outcomes and things to measure? And based on what evidence?" Lou Brooks: I think that it also has the potential for driving the R&D equation. There have been over the last, let's say 20 years, a lot of products, a lot of therapies that don't have a proven significant difference from existing therapies in the market. And by looking at the value-based agreements, I think it opens the opportunity for organizations to take a harder look at where it makes sense to invest. Does it make sense to come out with another diabetes therapy that might, might address a certain small portion of the population? Or is it better to invest in a gene therapy that is likely to have a broader impact in care in a rare disease? I think those are the types of things that the evidence and looking at the right questions and uniting life sciences companies with these other stakeholders, open up the opportunity to reevaluate and cascade through the rest of the process. Rae Woods: And that's because you're saying that only when we actually come together and only when we're actually generating evidence in new, different kind of unique ways, will we be able to understand how to map therapeutics of all kinds, whether it's a device, a drug or otherwise, to the specific person that is going to have a better outcome? Or the population segment that is going to have a better outcome, which is of course, one of the major goals of value-based care. Am I tracking with this so far? Solomon Banjo: Yes. Lou Brooks: Yes. Rae Woods: Now, allow me to be a pessimist for a moment. Half of the value-based care equation is the value to the patient, is the better outcomes, better clinical quality that might come from a specific drug going to Rae that's going to be different than the drug that works for Solomon. But the other half of the value equation is cost. Is there a way that we can be thinking about evidence now to support the cost aspect of the value-based care equation? Pamela Divack: I would say now more than ever, because we know that a lot of the treatments that are in the pipeline right now might be really high cost. And you'll need that evidence to prove that they work over the long period of time that it might require to see those outcomes. So if you think about gene therapies, precision medicine, even some biosimilars, biomarker treatments, you might need to now collect real-world evidence over a longer period of time to prove they work and then to pay for those outcomes in the long term. Lou Brooks: Yeah. And that gets back to collecting the right data, asking the right questions. To Pam's point, if you've got a new gene therapy and you want to evaluate what its impact is, well, then you need to collect the genetic information on that particular patient. You need to collect the actual clinical outcomes for that patient, and the cost. Traditionally, we focused on the cost and other utilization metrics to evaluate therapies of old, but we have to change the way that we ask the question and collect the data, because these newer therapies claims data by itself isn't going to be enough to determine whether or not they're actually working. Rae Woods: And, I guess, if I can argue with myself, one of the other benefits of moving to value-based care is to reduce waste in the healthcare system. And I don't think I can think of anything more wasteful than throwing a treatment at somebody that isn't going to work. Especially if now we have evidence that it isn't going to work. Lou Brooks: That's going to take a fundamental shift in the way that we, as an industry, deliver medications. I mean, think about prior authorizations and utilization management programs that are run to kind of help guide the right drug to the right patient. Those aren't necessarily designed today to factor in diagnostic test results or genetic information. We're evolving. And there's a limited set of those that do exist. So Herceptin is a perfect example, where if you don't have the right BRCA expression in your gene, then the drug isn't going to work, so- Rae Woods: Yeah. So if you have the BRCA gene, made famous by a couple of celebrities, including Angelina Jolie. Lou Brooks: Correct. Then that's a perfect example, but is that scaled? I mean, is that going on everywhere? No. Are we moving our way slowly to get there? Absolutely. But if we truly want to kind of eliminate waste and change that equation, we need to change how we think about collecting the evidence on the diagnostic journey for these patients as well. Because it shouldn't be, I failed on these four drugs and now I can get drug X. Theoretically, it should be, I do this particular test. It tells me I've got the appropriate or don't have the appropriate mutation. And then that guides the next portion of that diagnostic pathway to an actual medication. This medication's appropriate, because you've got that modification. This one is not. Solomon Banjo: So one thing to sort of repeat myself and bring up something that's underlying a lot of the tensions we're hitting on, is the tension between clinical outcomes and data and financial realities. So we've talked a lot about cell and gene therapies where the value could accrue over a decade plus, and we have to do what Lou said, ask the right questions, collect the right data to prove that. But most health plans aren't thinking about their populations in a 10-year window. And so there's that tension of, oh, Rae is on my plan right now. I should pay for this thing that has a promising result. How do I then do so in a way that aligns with my financial realities, that in all likelihood in four years, probably sooner than that, Rae might not be on my plan? And so how do I make sure the value accrues to me for my upfront investment? Rae Woods: How do you ensure that the value accrues to them? I mean, I'm thinking about employers too, or some other are purchasers. Employers don't necessarily have an incentive to think about the very long-term health impact, because obviously, employees don't stick with their employer anymore for 10, 20, 30, 40 years, like they used to. Rae Woods: So while in theory, I can see how each part of the system needs to work together to make sure that we're delivering the most cost effective treatment to the right patient, and we're using this new kind of evidence to get there. There's still part of me that's hesitating about how individual stakeholders are going to see whether or not this is worth it. Payer, just being one example. Lou Brooks: It takes a third party to connect the dots for them. That's where working to build out novel data streams that connect payers' data over time so that you can truly track a patient longitudinally for five years, 10 years, or whatever the case may be, offers the ability then for that payer to sign that unique VBC. Because if the payout is over, let's say six years, say that's what the VBC is written on, if you build out the infrastructure from a data standpoint to track that patient, then all right, well, I know that patient. Okay, they were here with me for two years. They leave. They're with another payer for two years and then with a third payer for two years. Lou Brooks: But I've seen all of their information, because I've worked with an organization that's connected all of those dots together for that single patient. That allows you then to adjudicate that particular agreement and go, "Yes, the drug did work for patient X. It didn't work for patient Y. And here's what the adjusted payout looks like as a result of that." That is the only way, because to Solomon's point, payers really look at the average lifetime of what a patient is, which is about 30 months in the plan. And same thing with an employer. That might be a little longer. That might be four or five years. But the same general rule applies. They're not thinking about these treatments that might have a 10-year impact. And we have to get more creative in how we connect those dots to evaluate those agreements more effectively, because it is challenging otherwise. Rae Woods: And my understanding is that we're still fairly early on in thinking about these new forms of evidence. We saw a bit of an explosion in real-world evidence and real-world data because of the pandemic. But maybe that's the only example of a drastic increase in thinking about evidence in different ways. So as we ask better research questions, as we expand the diversity of clinical trials, as we start collecting more real-world data, I think we'll learn more about how outcomes are different for different people and different populations that may be within that 30-month window. We just don't know what we don't know. Lou Brooks: That is true. Solomon Banjo: We can know some of what we don't know. And this is why I think partnerships get interesting, because we've heard from a few health systems and physician groups that are now recognizing the potential value of their evidence that they're generating. And saying rather than, "Oh, we could run all of our own analyses," it's like, "What are the biggest questions that are impacting our strategic goals? And can we work with another partner on Optum Life Sciences to say, 'Hey, from my market, can you get me data to better understand what's going on with this population, where we see outcomes that we don't like or where we know which trying to make a significant change?'" And so using their real-world evidence and data to inform better questions. And then saying, "Hey, as a partner, Life Sciences, help me grapple with this." I think that's one of the ways people don't think about life sciences organizations and vendors in that space that they could be, is helping you answer those questions that impact your strategic imperatives, your clinical focus areas. Rae Woods: I want to keep thinking about each segment of the ecosystem, because we are rapidly heading towards this future state where there is a ton of new, more useful evidence that is out there. But of course, life sciences companies can't just tell the industry to start using this new data. That's not how this is going to work. So let's keep talking about providers. How can providers actually use new forms of evidence? And what can they do now to start preparing? Pamela Divack: When it comes to individual clinicians and how they make treatment decisions for patients in front of them, they're now looking through that real-world evidence to ask questions like, how might this treatment work with patients with X demographics or X comorbidities? So they're looking to that evidence and trying to incorporate it into the treatment decision in front of them. But as we're thinking about provider organizations overall, and Solomon alluded to this, a lot of them are trying to make sense of the data they have internally to generate evidence on their populations or their patients. Because one of the big criticisms we've heard over the last few years of that RCT evidence is, "Well, this looks nothing like the patients that I treat. This looks nothing like my population." And real-world evidence is getting us one step closer to actually saying, "Well, this is how it works in the subpopulation of patients that I treat." Rae Woods: But realistically, how can we make it easy for clinicians to use this new data and this new evidence? I'm imagining the pushback from the clinician who says, "It is hard enough to keep up with care standards today when there are care standards for everyone, not for a niche population or when it doesn't mean understanding how this therapy is going to impact a single human being different from the rest of the population." I mean, practically speaking, how do we make this easy for providers to actually use the new evidence that's being gathered? Pamela Divack: Right now, we're seeing a lot of provider organizations embed that evidence directly into the EHR, because that's where clinicians are going. But whether that's actually the right solution, I think we're at an inflection point. We do see clinicians turn to things like online clinician communities to stay up to date and see what their peers are saying or who is commenting on what latest evidence study. But keeping up with that, all this evidence, is definitely a challenge. And figuring out the right solutions is something the industry will need to converge around. Solomon Banjo: I'll take it a step further and be a little bit more cynical than Pam, where I'm like, I don't think people are asking that question enough, Rae. And so I'm not certain we're going to develop solutions that are really going to address this. When you think there's this famous study that gets quoted that medical knowledge doubles every 90 days. So you're telling me a med student doesn't even make it a semester without the body of medical knowledge doubling. Unless we think drastically different about this, think about how different tools, technologies, artificial intelligence can help the clinician, it's just untenable to just assume that they and journals and other rule-making bodies can stay up to date on everything that's going on. Rae Woods: Yeah. Lou, this speaks to another part of your expertise. Is this a world where we can see artificial intelligence and machine learning actually helping to ease the clinician burden here? Lou Brooks: Yes and no. The challenge is getting it implemented responsibly and accurately. There is the ability for that modeling to, just like any other modeling, to generate spurious results. So you have to be careful about how much control you release to a machine at this point. But ideally, that does hold the promise that we can eventually get there. And perhaps not necessarily in the way that we may be thinking about it today. It may be identifying areas for further drug development and research, or diagnostic research to help identify these populations on an ongoing basis. Because I think one of the challenges that we have today, to Solomon's earlier point about medical knowledge doubling every 90 days is, even if today we can identify this drug didn't work, this drug did for a particular patient, what we don't know is why. Lou Brooks: So it's difficult at this point in time to be able to do more than say, "All right, this patient, didn't work. There's these other patients that look similar to this patient. So, you know what? Maybe I won't try this drug as a first line therapy in those types of patients." But until we truly take that knowledge and get that into the hands of life science organizations, to get them to research that, to better understand the why, all we know is there's a pool of people it works for. And there's a pool of people that it doesn't. And that's not where we need to be. We need to make sure that those connections are made and those collaborations exist between stakeholders to ensure that we learn from that identified knowledge. Rae Woods: That's exactly right. And that is what Solomon was saying about the role of partnership. It is not just that providers need to take the new evidence and information and implement it in their decision making. They also need to understand that there's a feedback loop that goes the opposite way. That they play a role in collecting the data and in helping ask better research questions like, why does this treatment not work for this particular population? I want to come back to a different segment that we touched on, which is the health plans. How do they need to use new evidence? And what steps do you want to see insurers take today to prepare for a world where this new kind of evidence is the norm? Solomon Banjo: What a health plan often does, and it makes total sense, is say, "Hey, how as this new treatment compare to previous treatments? Is it better? Or is it worse? And let's use the same historic endpoints to measure it." As we think about the challenges we're facing, how the whole industry has evolved, I don't think we're asking enough, "Are those historic endpoints, the right endpoints? Do they give us the information, whether clinically or financially, to be making the decisions we need to make?" And so I think it's a very hard thing to say, but consider when those historical endpoints aren't serving you and be more open to embracing new endpoints, patient-reported outcomes, that can actually help you achieve your goals while granted yes, not making it as easy to compare to historic performance. Lou Brooks: But Solomon, to your point, I think the challenge is that we are, depending on your lens as a stakeholder, we're so entrenched in those historical kind of norms or practices that that's where there's a huge amount of work yet to do on getting organizations in one part of the healthcare system to look through the lens of someone else and really evaluate what those data elements and that evidence actually means for them, rather than reverting back to the standard norms. And I'd be interested to get your perspectives on how do we help shift that mentality? Pamela Divack: One other norm, to be challenged here with all this evidence emerging is also how frequently we revisit decisions. Because with so much new evidence coming out, we might be able to take a more dynamic approach to formularies, or guidelines, or all of those decisions that were a little bit more static in the past. So I think that's one way payers and also providers can be using this new evidence. Rae Woods: Well, Lou, Pam, Solomon, I want to thank you so much for coming on Radio Advisory. Before I let you go, I do want to give each of you the chance to speak directly to our listeners. And I'm going to give you a bit of a choice here. You're going to be giving one key takeaway for our audience. And you can choose which stakeholder you want to speak to. You can choose to continue to speak to life sciences, or to payers, or providers, or to the industry as a whole. So no pressure. But when it comes to the state of evidence, what's the one thing that you want our listeners to know? Pam, let's start with you. Pamela Divack: I'll start by speaking to life sciences leaders. I think it's clear how real-world evidence can be really critical across all parts of the product life cycle, whether it's an R&D and target selection, clinical trials, patient recruitment, even post-launch interactions with customers. But you need to think about it strategically. So we're seeing organizations figure out, how do we infuse RWE and evidence in everything that we do? How do we organize internally? And I would think it's important to keep elevating RWE to be a strategic imperative. Rae Woods: Solomon, what's your takeaway? Solomon Banjo: My takeaway would probably build on Pam's, so focus on the life sciences industry, which is like any industry in any large organizations, there are lots of silos within it. And we've sort of been painting over that. And Pam hit on the fact that I think if we're going to maximize the ability for life sciences organizations to be true partners, to really move forward the field, they need to think about how they break down silos when it comes to evidence. So having evidence and real-world evidence, not just be housed in the [inaudible] function and sort of the traditional places. Or when we think about the value of diversifying clinical trials, how do we think about it holistically and not just through a clinical lens, a commercial lens, et cetera? Solomon Banjo: Because when you take that step back and say, "Oh, how would this help us as an organization?" You actually realize that there's tremendous benefit in that example, to being able to tell physicians, payers, "Hey, this reflects the population impacted by this disease. Here is how we do or do not see differential impact in the positive or negative. Let's have a conversation around value that is more robust as a result of that." Rae Woods: Lou, what's your final message? Lou Brooks: I'm going to diverge from my... I'm going to send this out to everybody. In my mind, it's innovate. Don't be stuck in the old models of evidence generation that we've had to date. Think about what you'd like to have and then work towards that particular goal, rather than continuing down the same paths that we have been going down for a number of years. If we truly want to better manage costs, improve patients' health, then we have to think about the models a little bit differently. We have to change what we're doing in order to get there. And it's going to take time. And it's not going to be easy. And we're going to stumble along the way. But it's the only way we're going to make a difference. Otherwise, we'll just continue down the same road that we're on and never really get evidence and value where we really need it to be. Rae Woods: Nailed it. Way to have the mic drop moment at the end there, Lou. Huge thanks to you, Pam, Solomon, Lou, for coming on Radio Advisory. We are going to have to chat again. Pamela Divack: Thank you. Solomon Banjo: Thanks, Rae. Lou Brooks: Anytime. Happy to do it. Rae Woods: The world of evidence is rapidly changing. We've come to terms with the fact that our historic way of gathering evidence was actually flawed. So we're adding more diversity into clinical trials. We're asking better research questions. And we're looking to real-world evidence to ensure that our drugs and our devices create the outcomes that patients need and payers want. To make this the norm, we're going to have to work together. So remember, as always, we're here to help.