Event 1 - Evidence Generation: Quality of data for decision-making

Prof Carole Longson
Eric Low
Dr. Dan Ollendorf
Dr. Wija Oortwijn

Guest Experts: Dr. Frank Hulstaert and Prof Ian Tannock

Ian Tannock: The problems stem from the fact that almost all of the large trials now are sponsored by pharmaceutical companies and clearly their aim is to make a profit. Now, sometimes the goals of making a profit and improving outcomes for patients are coherent, but sometimes they're not. I think the problem is that there's not enough control over the way that the trials are designed. Chris booth recently wrote a paper which reviewed modern randomized trials. Now some 90% of them are sponsored by pharma. The government hasn't put the money or doesn't put the money into doing independent trials or extremely rarely. In those trials there are many of them that have inappropriate comparators. So if you use a comparator where the outcomes are relatively poor, and sometimes that's done on the excuse, this is what's used in the real world rather than the best evidence from previous trials, which is absolutely nonsense because if you say the patients are well enough to get the most recent treatment that may be more toxic, then they're not well enough to get the new agent that's being tested, which is often a more toxic.

The endpoints are problems: 80% of the end points for trials that are put before registration agencies, like the FDA and EMA, instead of using overall survival and quality of life, which are the ultimate things we want to improve for patients, they use progression- free survival. We were recently working with a statistician who showed that because of the phenomenon now of informative censoring, where people who drop out before reaching the end point of tumour progression, that you can get an apparent benefit from a drug that is totally inactive, but toxic. And that is very worrying. There is a move towards accelerated approval for breakthrough drugs. So I'm not saying that there aren't some very good improvements in treatment, some of the immunotherapy, for example, but it's been extended quite widely. And although the FDA may require a subsequent randomized trial to be done in general, when that hasn't been done, the drugs have not been withdrawn.

Pharma has done some good things. The development of the COVID vaccines is a good example. But on the other hand, only half of the current anticancer drugs out there have been shown to improve survival or quality of life. Quality of life is often very poorly done. And when they have done that, a lot of them have only done it in by a trivial amount. We have a lot of problems and a lot of wastage and rather inferior drugs that have not been adequately tested.

Frank Hulstaert: This is very much the case also here in Europe. The clinicians needs are the same needs as those of the health technology assessment bodies. And they want to see

competitive data and in order to practice evidence-based medicine. So indeed the patient relevant population to be studied, that reflects also the target population that will be reimbursed is critical to have it in the trial. On the intervention, immunotherapy for instance, one aspect that we see less and less also with the biologicals is that the dose finding and the duration of treatment is not always well investigated when it comes to the market. And that is of course also impacting the budget for the healthcare payers. That's also an important aspect of the intervention. Comparators among the HTA bodies of course, there needs to be alignment when industry is advised for the pivotal trials, so that there cannot be any discussion on that and I think that is possible.

In terms of outcomes, these surrogates that have sometimes already proven to be not valid as a predictor of overall survival but are still being used and that is a pity. One aspect is the start of such a competitive data. If we are now pushed into the post-market setting to generate competitive data, the problem is that the recruitment is done competing with routine prescription and it's very difficult and delayed sometimes when results become available.

The point would be to have these competitive data or these competitive trials at least started in the pre-market phase so that for the patients and the clinicians, the evidence becomes available much more early and that evidence-based medicine can be practiced more early and patients also then have more rapid access, not only to innovation, but evidence-based innovation.

Leeza Osipenko: We deal with an interesting issue of speed and the pressure comes from a lot of points in the system from patients from clinicians, from industry. And obviously agencies are under a significant pressure to make decisions sooner and on the evidence that they have available. For example, Frank, you brought an example saying we should start these studies, comparative studies, preregistration. Risks are too high. We need to guarantee that there will be a license. There are a lot of conflicting points here.

How do we balance the speed with requirements that we would like to have?

Wija Oortwijn: The alignment with the regulatory authorities. One of the issues is also that the regulators have different objectives compared to HTA agencies. They focus on the benefit risk profile while HTA are focusing more on the deep clinically relevant long-term effects, but also in terms of the evidence that industry is providing with regard to safety. The way regulators evaluate safety is different from what HTA agency do. That is also potentially a problem and of course, the industry might target market approval by focusing on the evidence bars that are set by regulators as well. The HTA agencies need different evidence to inform decision-making. Also the regulators can or have the authority to determine the quality in the size of pre-approval clinical evidence. they can also ask for post-approval trials. But they have the authority, not the HTA agencies. there should be more alignment with the regulators, coordinate better when they set the post-approval recommendations and it should also be done P market. The optimization should be in the whole line. There was also some uncertainty that the authorities have to deal with. They probably also need more evidence. So why not align the uncertainties that regulators have to deal with and the ones

HTA agencies have to deal with? That could potentially be one option. There is also a joint scientific advice that could potentially be an option to align more with the processes.

Finally, we are dealing with a technology push from the industry and with policy relevant question from HTA agencies, and these are often not aligned. We should also be much more focusing on how to identify technologies and how to set the priorities for health technology assessment.

Dan Ollendorf: The concern is that regulators do have standards and they differ in some respects from the expectations that HTA bodies will have. But as point of first principle, I'd like the regulators to live up to those standards just a bit more. Accelerated approvals are significant not only for oncology drugs, but for drugs and other spaces as well. There was a recent publication researchers at Harvard where Hussein Nazi from the London school of economics published a review of 10 oncology drugs that received accelerated approvals from the FDA. All 10 of these drugs over a three-year period, failed to demonstrate an overall survival benefit in confirmatory trials. Only a small fraction of these have been voluntarily withdrawn from the market. And in some cases, the oncology drugs advisory committee to the FDA recommended against withdrawal because of other considerations, no other options on the market, et cetera. Why does there even need to be a vote? If the FDA is saying, these are the conditions under which we'll grant an accelerated approval. And if you fail to live up to those conditions, why does there even need to be further discussion? Withdrawals should be essentially automatic at that point.

Another concern that was described in the problem statement is that oftentimes whether it's accelerated or not, when we see approvals come in for multiple technologies that are intended to target the same condition, sometimes HTAs hands are tied. There are subtle differences in trial design, differences in measurement, differences in timing of that measurement, that preclude even an indirect comparison of these new technologies. Which is unfortunate and also something that could have been avoided if the advice given by regulators to technology developers is consistent and taking advantage of some of the joint advice programs.

There is an opportunity to have those conversations early enough so that the evidence meets some minimum threshold for evaluation. There may still be challenges. We may have uncontrolled studies because it's unethical to think of a control group in certain trial designs. Maybe other efforts will have to be made to mitigate the problem in those circumstances. There are ways we can shore up the evidence-based now, but those require, essentially pre-approval and even pre-submission conversations.

Carole Longson: This problem is not new. Many of us have been around in health technology assessment for a very long time and it might've been magnified as the introduction of these accelerated processes have occurred. Many of us have sat around decision -making tables or observed decision-makers in HTA, look at new technologies and come to the realization that what evidence they have is poor for decision-making. And actually have we moved? It may not have in the actual delivery of the evidence that is needed. The mechanisms by which that could be achieved are in place now and they never

used to be. We've already talked about parallel advice that wasn't around 20 years ago, it came in five, six years ago. It is not being used in the way that it should, but at least HTA experts have realized that they need to get into that regulatory space as early as they possibly can in order to ensure that at least the right conversations are happening at that point. A good start could do a huge amount more.

The second thing is the realization of some of the regulatory mechanisms that it is not sufficient for them to just land a product on a market without at least exposing some of the uncertainties that can then at least be addressed. With the European perspective, the EMA have done a lot at increasing the transparency of their decision-making and communicating between themselves and HTA bodies to handover more effectively than they used to some of those issues.

Eric Low: There's a lot of complexity in the system. If you mapped out, from bench to bedside, the different stakeholders that are involved or have a stake in discovering, developing regulating, evaluating, appraising, and optimizing treatments, there are multiple. They're operating in an ecosystem that is completely disconnected. And what you see is different vested interests, so different stakeholders want to get different things out of the system. They've got different drivers, they're incentivized in different ways and they're rewarded in different ways. As long as that's in place it's difficult to move the dial. Part of it is around how do we get the right incentives in place, everybody to play differently so that each stakeholder wins, whatever that might mean, but that win is sort of unified around a common goal. And that common goal should be to get properly transformative, safe, and effective treatments to patients. And why I mentioned the sort of technology push and that's absolutely right we will exist in a push environment. Downstream from the laboratory, we react to things that come through the system. It comes through the system in a random way, in different ways. NASA took that approach to put an amount on the moon, it would never have happened. And yet we have rules in the system that allow this kind of thing to happen, where we don't have a demand side. So part of what we have to do is to think about where does the pool come from and is that pool voice evidence-based, is it unified and who is that voice from? And as I guess, the patient representative and we're all patient representatives on the panel because ultimately, patient-centricity is not something that just sits with the advocate. It's incumbent upon all of us in the system to think about what we're doing for patients. But first and foremost, we need to understand what's a patient wants.

We mentioned benefit risk as being something that's for the regulator, but actually when you try to work out what the value of a medicine is you have to understand, what does value mean from a patient perspective? What magnitude of benefit does a new treatment bring? And what is the probability of getting that benefit? Is it 1 in 10, 1 in 5? And where's the risk, and what's the probability of getting there? Is the risk serious and the probability of getting it huge? And then the value propositions very low? That's not what the regulator does and in HTA we're not really putting that type of data into the decision-making. So I think part of the challenge is around how do we connect up the system to get a balance, how do we get the right alignment in terms of interests? What currency that will cause the change? And it's patient data because research doesn't happen without patients.

We've got to get the patient perspective into the argument and patients or patient groups need to be saying that clinical trial is not ethical. You cannot do a single arm phase two study in oncology with under a hundred patients with surrogate end points, and six month follow ups. That's not acceptable. Until patients actually start and patient groups and patient advocates start to preach by and see that this is not acceptable, it's hard to see how things will change.

Leeza Osipenko: Is the patient perspective taken into account when evaluating surrogate end points which are used by the industry in the trials?

Eric Low: It's like in every therapeutic area, you couldn't get a super responder. And there's always an example of a trial or a clinical program in which the patient voice has made a big difference. I've been a patient advocate now for nearly 30 years and I can tell you I've contributed so much over two, three decades, and I don't believe anything I've ever said has materially impacted on either the design of a commercial study, and let's not forget about academic studies because that is as much a car crash, as commercial studies. I don't believe I've had any influence at all, and I would consider myself to be sort of an hourly smartly person. I'm not particularly good nor bad, but pretty good. And I haven't had an impact at all.

Leeza Osipenko: Any more specific comments on patient involvement and evaluation of surrogate end points?

Ian Tannock: There’s a lot of misunderstanding. Patients like to be told that the tumour is not progressing. Now, if that means that over a fairly prolonged period, the patient is not experiencing new symptoms of their cancer, that's important. But the problem is that also all new drugs, that toxicity, and clearly the value of an endpoint like that really comes down between a balance of improvements in quality of life because of control of symptoms and deterioration of quality of life, because what the drug does, and that is often quite severe and negative. And I think to answer that question, you really need to look at the quality of the survivor. And that may be different in different diseases. About the question of value of treatments, the larger organizations, particularly the European society of medical oncology (ESMO), has developed a magnitude of clinical benefits scale. ASCA also flirted with this, but it's not really pushed it since. But ESMO has continued to do so. They can look at the results of their clinical trial and set some criteria for what is considered a valuable treatment. And it relates to changes in survival, changes in quality of life and toxicity. They've sent out a formalized way of doing that. And if we could persuade the regulators to look at a criterium of value for drug registration, rather than a statistical test, which is not very meaningful. And the company has got around statistical tests by doing larger and larger trials to show that smaller and smaller differences are significant. Unfortunately, again, a study that I was somewhat involved in with Chris booth and colleagues looked at the design of randomized trials as to what proportion of them, were actually designed to detect or rule out a level of value as per the ESMO magnitude clinical benefit scale would be regarded as worthwhile. The answer comes out that only about 30% of trials are even being designed to detect or rule out a benefit that was considered valuable in that system. Clearly, we need to push pharma to design trials that will actually either show or value or prove that there isn't

sufficient evidence of value and they'll only do that if the registration agencies push them to do it. Oncologists and HTA people need to be pushing people to do that.

Carole Longson: Two points to make which are absolutely spot on for an HTA audience, talking about ineffective value and incremental effectiveness is exactly what we need. Unfortunately not what we get very often in those trials and all routes into this working better our systems approaches and through the regulator, working together with the regulator to then specify signal instruct, I don't know which one is best, which works best. Those that are developing the trials, actually, it wouldn't matter whether or not it's a company or an academic to do those in the best possible way.

We really need to recognize that certainly my view and I hope other HTA experts’ that the perspective we need to take is what is meaningful to those who are intending to benefit from the intervention. That's the perspective we need to take. Yet everything that is done in clinical development does not take that perspective at all. Even when we talk about good endpoints, they're still end points that have been developed without the input of patients. It's still as difficult as that to get that at least that endorsement from a patient perspective, that what is being measured is something that they genuinely wish to see. We need to do much more all of us in ensuring that somewhere in the system, the call for the patient perspective, in the development of new clinical endpoints, clinical trials design, and definitely registration trials, which the regulators have now recognized, but just keep the call going. I think it would be an extremely important developments and at least it would reinforce the reason that we're all doing this, which is the aim to get beneficial treatments into the hands of those that would most benefit from them in a way that's sustainable for a healthcare system. That's the goal.

Dan Ollendorf: Sticking with oncology for the moment, one of the conversations that I'm hearkening back to during my time at ICER was with the cancer support community, which is a relatively large advocacy group here in the US that has a research focus. They have been sounding a drum beat for many years about how these studies inadequately, if at all capture the low-grade daily toxicities that these cancer patients have to deal with. So there might be good measurement of overall survival. There might've been even be good measurement using standardized quality of life instruments, but those do not capture the impact that these cancers have on the patient's daily life in total. And again, it's a situation where they've been talking about this for a long time. People nod their heads sympathetically, and then nothing ever changes in terms of measurement.

One of the other things around this notion of saying, ‘in this particular situation with this particular cohort of patients, we find this proposed study design completely unacceptable’. That seems to be a place where HTA and the patient community can join forces and advocate together. Now, in my connection to the broader international HTA community, it's not as though we are a poster child for good and consistent patient engagement. We have a lot of improvements to make there as well, but there's an opportunity here. I wrote a blog piece essentially kind of registering my shock at the approach that in this case it was the US regulator took to evaluating gene therapy for haemophilia. This is a condition that we've known about for hundreds of years, we know what the patient important and statistically

important measures of outcome are. We evaluated as a community, Emicizumab an antibody treatment for haemophilia a couple of years ago on the basis of a reasonably well conducted largish randomized controlled trial. The next available treatment for haemophilia was a gene therapy that the FDA decided to evaluate on the basis of a phase I-II safety study of 15 patients, uncontrolled and was designed essentially to measure factor eight levels rather than patient important outcomes like joint bleeds and hospitalizations, which makes no sense. We know the size of the patient population. We know the potential cohort of enrolees. We don't need to limit ourselves to a study design like that. We can demand the kind of study design that will give us the answers we want.

Wija Oortwijn: Of course, we need to look at the patient relevant outcomes. We were talking about the role of industry, but we should also talk about the role of the clinician because they are also a part of the clinical trial. We as HTA community are also not well aligned with the clinical societies in itself. How can we join forces between the communities, the patients, the clinicians, and then the HTA community? And also with regard to the parallel scientific advisory or scientific advice in general, because it seems to be that there is scientific advice, non-regulatory scientific advice and there is regulatory scientific advice. In the UK at least is that for NICE, patients and clinicians are working together in that particular scientific advice. But I'm not quite sure about other countries if we have the same, we don't have the same in the Netherlands, for example. There is some movement, but still it's very small and it's of course, very targeted to the UK market. So can we broaden that scope?

Eric Low: I've often advocated to patient groups and we should see HTA as the friend because we can help each other out by strategically aligning on that demand side. There's absolutely no question about it. Also, Dan mentioned the trial, it may not be one study. It may not be one trial that is sufficient to generate the data that the decision-makers in the system needs. We need strategic evidence development and that may mean aligning the research efforts of health systems and clinicians to those of industry. So that a well-designed UK based phase 2 B study, for example, might be sufficient to mitigate against the uncertainties, even over well-designed appropriate regulatory study. And that would be a great way to bring the clinical and research environment networks into the clinical development program. It's good for our country because of inward investment. And the scientific advice should be available for academic studies as well. The package that's ultimately submitted should be a combination of the regulatory data, local data from academic studies, well-designed studies, all of which has been run through scientific advice. That would be one way to do it. And in a past life at myeloma UK, we designed an academic clinical trial network that was set up to partner with industry to share that burden of evidence development through academic hybrid studies. And we pioneered a scientific advice product with Leeza's help through NICE where the company would take their regulatory data and the data from the academic clinical trial and present it as a package. That's just one example of where some of the issues in the system can be mitigated by aligning efforts of clinicians, patients, the regulator, the HTA body and industry.

Ian Tannock: Two points. In terms of liaisons between HTA and value systems I actually think that the people I know at ESMO who have developed the clinical value scales led by Nathan Sherman in Israel, already have these liaisons with some of you,these are very open-

minded people and they would welcome that type of liaison because I would love to influence the regulators and the HTA agencies. The clinical value becomes established as a criteria on which drugs are first approved and then made available, funded in different countries. I'm sure they would be enthusiastic about doing that. And if that happened, it would have a secondary effect because it would push the pharmaceutical companies to recognize that they have to meet that criteria.

The other thing is about scientific advice. I have sat occasionally on committees put together by companies to give advice about trials. But clinicians who take part in those, a lot of them are being paid quite handsomely. I personally don't accept funds from industry, but a lot of them are, and there is if you stop to sort of push for things like value systems or end points. When it comes down to it, the company are going to get their drug to market as quickly as they possibly can. And if the FDA and the EMA say, you can do that with an end point by progression free survival instead of overall survival, that probably cuts about a year of the length of the treatment and that's a year gain before the drug goes off patent. And that's the way the industry people think. And to be honest, many of the physicians see that as well. I can't really influence that because that's the way things are. They need to separate, 'let's get this drug to market as quickly as possible' to 'is this the appropriate product to get to market with the appropriate evidence?'. And I think we need more independent panels. Personally I think that people sitting on FDA advisory panels and NCCN advisory panels should be forbidden from having funding from pharmaceutical companies. I wrote an article some time ago, an editorial about purchasing silence. And this is what the companies are doing, they're purchasing silence, unfortunately many of my colleagues by giving them handsome honorarium.

Carole Longson: I don't think that HTA advisers would have any problem whatsoever selling companies that they need to do comparative incremental effectiveness trials. The issue is how often do you have that perspective actually in the advisory discussion either in industry sponsored advisory panels or in the formal scientific advice processes? Isn't it happening often enough? And the way that companies tend to use HTA advisers is much more on the commercial value proposition than the clinical value proposition and that's unfortunate because there's so much to talk about on the clinical value proposition and there's no point going to the commercial value proposition until you've had that conversation.

The second thing is that the evidence is never going to be sufficient for an HTA decision, mainly because of the timing that an HTA decision needs to be made which is close to regulatory approval. You've never had the information you need to get a full perspective of value and incremental effectiveness and also because HTA generally takes a lifetime horizon, we want to look at the full course of the condition over time and the impact of new technologies on that. You're in a circumstance where you've never will have that evidence-based. We know that the evidence-based that can be presented at the point of decision could be much better than it is now talks about that. But the other thing that perhaps we haven't explored yet is what technical tools and what scientific tools we can deploy to actually bridge and maximize the value of the evidence that we do have. From an HTA perspective, the use of those tools has increased over time and has been useful.

Just an example, it's been brought up before the relationship between progression-free survival and overall survival. There are ways to be able to model that in an evidence-based way. We know that in different cancers, that relationship is not the same. So you can't generalize from one cancer to another, even for the same products, but there are techniques that would allow you as a decision maker to feel comfortable in using PFS. In certain circumstances in certain conditions. And so the sophistication of the methodology, if I could put it that way of HTA can help to some degree to mitigate the issues that we've that we've highlighted and explored. It doesn't take the needs to do everything that we've already discussed.

Frank Hulstaert: On clinicians going together with patients and HTA for advice I think that synergy should definitely be used because what we see now is if there is parallel advice that the regulators advice is followed and the HTA advisors not followed by the companies for many reasons why would a company do this comparative trial, but that is more costly and more risky?

Finally, because you risk showing that you don't have a benefit. The problem with the post marketing situation is that at that moment, even as we heard, when you showed, there is no efficacy in the post-market phase, it's very frustrating for all the parties to withdraw a product from the reimbursement from the market. So you're in a different setting when you are in the post market compared to the pre-market setting. And that is why it is so crucial to have that full package as full as possible already in the pre-market phase, because it's a missed opportunity if you postpone it to the post-market phase. That is important to take into account. And if it is in the post-market phase, then it should be a research use only situation where only patients entering a large, pragmatic randomized trial or so are reimbursed. And I think that we should think about incentives also for industry and co-financing because indeed the trials that take longer the healthcare payers have also an interest to have the evidence developed. We need to have a more helicopter view to see the full picture from the start of the clinical development to an evidence-based decision that can reach the patient. And that should be in a period that is as short as possible, independent of the parties involved and that's where we should start that.

Leeza Osipenko: You suggested many solutions and express different wishful thinking scenarios of what the advice should be: what, from your perspective, are the current key bottlenecks? And how would you reflect on them? Are we dealing with the resource problems? Are we dealing with the designation problem where each agency says that's not my remit? Do we not have enough authority in the system? Do we not have enough drive? Because I think there's plenty of understanding in the system. There is a lot of theoretical knowledge, but we have this trouble getting things into practice and one excellent example about managed access agreements where the idea is absolutely brilliant, the way it's been proposed to say we have to get drugs to the market quickly, but let's create a system where we generate this evidence where we get the comparative evidence. As a result it's not really working because many poorly performing drugs that probably should not be on the market or at least not cost more than they are costing compared to other therapies, continue to be priced at a premium and remain on the market because we're not fulfilling those comparative studies. We're doing it 20 years later, this evidence comes up way too late.

Dan Ollendorf: When early advice is offered innovators will focus on what the regulators say and not necessarily adhere to what HTA say. The bottleneck is either lack of standard setting or lack of authority to enforce those standards that are set. HTA bodies that offer early advice do it on a voluntary basis, but it's not compulsory. But as part of that voluntary agreement, HTA could make it clear to the manufacturer that their advice will be generated from their expectations, their internal reference case for what they want to see when it comes time to make a submission. And if what ends up coming in, violates the reference case, then manufacturer needs to know that's not going to end well for them. I think that HTA, hopefully in concert with the patient community could develop a set of standards, even a set of standards that is condition or disease specific to say, this is what we expect for evidence. And if something will depart from that, we need a very good reason why.

Eric Low: One of the challenges, at least from a UK perspective, we've got is the pressure on each team. Decision-making often comes from government and administers and it's mostly an industrial strategy. Therefore it's very difficult for NICE to say no. There's all kinds of flexibilities in the system. There's no question there's some upsides, but there's downsides as well because it de-incentivize the suppliers to bring better data to bring better inputs, because we know at least in the UK, NICE, there'll be some sort of lever or some sort of something to help get them over the line. It's a bit like marking your homework. In the context of managed access, it should be reserved for the datasets that fit the reference case that the company's demonstrated we've done everything possible to mitigate against uncertainty and the uncertainty there is just a fact and therefore it would qualify for managed access. But managed access should not be used for a data set that's been mishandled and mis designed because it's like encouraging our kids to misbehave all the time. It doesn't work.

Frank Hulstaert: On managed entry agreements, I think they are great to control the budget, but not to generate evidence. And some people think that just giving a non-public discount will create effectiveness. Some people really think that. And we need to convince them that is not the case. Effectiveness is not changed just by the price.

Carole Longson: If you're thinking about evidence generation in conditions of uncertainty, that uncertainty actually does have a value. And it can be that value discussion can be deployed on the price side. With agencies that are looking at clinical and cost effectiveness, the value of collecting information, additional information can be offset against the willingness to pay threshold. I'm not taking my sort of pure evidence Perspective here, because you would always wish to reduce that uncertainty with further evidence. But from a decision-making perspective in a reimbursement situation, you can mitigate those two things.

What I wanted to touch on was your challenge. What are the bottlenecks? I've got three, and we've already touched on one with other panel members and that's incentives and disincentives to do the right thing. I don't think we are sophisticated as an HTA community in that regard. I think standard regulatory processes are much more sophisticated in that way. So I think we could do better there.

The other two issues bottlenecks are alignment and timing. With the timing, we absolutely need to be having conversations about evidence generation earlier than we currently do. We just need to find ways of being able to do that. Scientific advice is one mechanism at the moment. It's not the only one. Probably we should be trying to get our HTA advisers on the advisory panels that companies are using because that conversation will then happen. It may not be taken up, but at least it's going to happen. At the alignment I think the biggest change that could happen and it is I'm very hopeful, certainly in the UK that there is the prospect of this, is that a regulator and an HTA body sits together and decide and design what they are going to say to companies. And that is then a jointly endorsed advice process, and that needs to be taken up through the regulatory process. And the regulators are also checking that as well. And then when it comes on to the health technology assessment, hopefully both the evidence is there, but also the alignment between the regulatory and the HTA body has already been in a sense baked in to the clinical development process.

Leeza Osipenko: My biggest question to this is what is the driver and the push for this? What can make this obligatory? Because we've been having this system in place for a while, and the problem is that it's advisory, it's an option. And I think it gave us amazing practice to come up with all these solutions, but we don't have a compulsory system, so to speak in place, to clean it up, to really make this work for everything, not for three active companies who decided to approach the process. So we need to think of the trigger who can do this actually.

Eric Low: I think if the demand side is weak, the system won't work. And I think it's incumbent upon patient advocacy groups and clinicians and nurses, in any therapeutic area, to map out an investment? To understand what do they want, what do they need, what are the red lines? How do they see the treatment pathway in the next 10 years? And we've got a bunch of patient advocacy groups and medical research charities funding billions of research that's pointless and meaningless. They should be investing in methodologies and projects to prioritize research, to understand what the most important clinical questions are that need answered either by industry or academia. And it's a bit like, and maybe this is not a good analogy: people build an aircraft carriers and helicopters and rocking up to the ministry of defense and knock on the door saying, 'we've built a helicopter, do you want to buy it?' I think we do that badly. Write a blueprint of "this is what we need and a helicopter, we need it to do this, do that". The next thing, go out with the tender and the best company designs that helicopter. it's 10 years late, and it's six times more expensive, not normally, but it's that kind of conceptual mind shift that we need. The conditions of patients can drive to better understand what we need and give that tender to industry and say, "we need these drugs and if you get these drugs to us, that pool is going to be there". And that to me is a massive bottleneck. And just on managed access and risks, because I think people have made some very good points about that. It's to manage the financial risk, the unknowns of the drug, they don't go because of that. But we're transferring that risk onto patients who are often making decisions about treatment under very difficult circumstances. That risk doesn't disappear. And we have no evidence really on what level of risk is acceptable to patients. We assume that every patient will do whatever it takes to survive three more weeks or three more months. It's absolutely not true. People will trade quality for duration. And we just need to reflect on how we would advise our loved ones,

say with cancer in these situations, we would not be saying, 'take that next treatment'. We'd be saying, 'look, weigh up the pros and cons, you know, do you want good death or bad death? How do you want to die effectively?' We've got no evidence for any of these. And until we get that, I don't think we're going to get the change in the system that we need.

Wija Oortwijn: First the alignment with the regulatory process. I think we should be focusing on identifying the uncertainties that both the regulatory agencies have and the HTA bodies have and discussing them indeed together what the necessary evidence would be. And then also the timing for the generation of that evidence. And this should not be indeed voluntary, but mandatory. And I'm also wondering if the EMA and the FDA could work together in that respect.

But before that, I think the issue mentioned both by several analysts is the incentives. We should not forget that we have European regulation for drugs, for orphan drugs, for paediatric medicines. And these incentives, the regulation has been evaluated. I've evaluated the regulation myself. And there are very uneven incentives into regulation. And of course, European commission tries to circumvent or withdraw or make it different, but it's very difficult. And I think that's one thing because there are incentives for industry to make these drugs and there are, indeed, very little incentives for academics to get money for research in areas where there is high unmet need. 'what are the needs?' And you could look at European perspective of countries in Europe, but of course we also have other countries. We are not talking about Europe. But our epidemiology profile and needs is completely different from partly the US and Latin America Asia Africa. Policymakers should be more aware of this. COVID clearly showed that there was a need for vaccines. And of course, then there was a huge push and of course industry was interested because there was some gain to be win. I don't know how we can, as an HTA community, because this is a systems discussion and it involves the policy makers at the European level at least, align research priorities, industry policies, and of course access to care. And to me, it's a huge issue that we as HTA community, we can influence, but we do not have the authority. And that's the problem. We are sometimes at the sideline and that is frustrating, but at least we can put it on the agenda.

Frank Hulstaert: I think we are in a special time now, because in Europe at least we have this HTA regulation starting. We have the pharmaceutical framework being started. I think it's the right time to have the European commission also involved as well as the member state governments to gather the HTA bodies with clinicians, the scientific societies and the patient organizations to indeed ask for more competitive evidence already in the pre-market situation.

And one point that has not been discussed yet is the possibility of having platform trials to generate this comparative evidence. Because by definition then you also have a harmonization of the trial end points in this platform trial like the companion diagnostics for the immunotherapy. That thing went wrong because every company selected another companion diagnostic and then the routine care, it's kind of a mess now.

Dan Ollendorf: I'm trying to link both the post decision environment with managed entry and managed access and the early advice environment. I wonder if some principles around this notion of war gaming, it's not the greatest term, but you'll see what I'm trying to say, might be set forth, meaning manufacturer could describe a couple of scenarios under which this compound in development might be approved. One of which could be a traditional regulatory timetable and other could involve some sort of accelerated approval. And then expectations could be set for those multiple scenarios around what HTA would want to or need to see. So for something accelerate. And some of that evidence will have to be generated in the post decision space, but there could be guardrails put around that. But it seems to me that there's some interesting guardrails that have been set around that with a time limit on the generation of additional evidence if that evidence is not generated, then the product essentially is disinvested. Contrast that with leaving those expectations in the hands of the regulator, you have the approval by the FDA, in the US with an expectation that the manufacturer could produce confirmatory evidence within nine years, a trial will take 18 to 24 months at most to conduct and produce those confirmatory endpoints. If things stay the way they are with those kinds of loose expectations, then we won't get what we need. And as Frank said, managed access will turn into a cost cutting or cost management exercise or will stay that way for the most part.

Event 1 - Evidence Generation: Quality of data for decision-making

Search