 So what I'll talk about today is whether the EU incentive for drug repositioning has been effective. Fortunately, Trevor Cook has already introduced the IP that I want to analyse today. He described the 8 plus 2 plus 1 formula. And what I'll focus on is the plus 1, which is a specific type of regulatory IP for incentivising drug repositioning. I'll touch more on the specific details of that formula in a moment. The big idea that I really want to draw your attention to is that the plus 1 commenced in November 2005. And I've analysed all the drugs that have been authorised and repositioned since the EMA began in 1995. This means I have a data set of repositioned drugs before and after the exclusivity began. Which means we can judge in a way whether it's been effective. So only a handful of people have seen this data and I want to engage you all in really working out what that means. I warn you now too that like Matteo, I'll ask for some audience participation. But let's see, I expect maybe there'll be some more responses than our students. Right, so the plan for today, first of all I'll go into some definitions that are the key to really the method that I'll apply. I'll look at the incentives for drug repositioning and I'll look at what's called the repositioning problem. I'll also look at some procedural aspects and historical notes on the plus 1 that I think you'll find very interesting. In the methods of results section I'm going to work towards answering two questions. One, has the plus 1 increased the proportion of repositioned drugs? And then secondly, has the plus 1 increased the number of times drugs are repositioned? Because some drugs as we'll see are actually repositioned quite a number of times. In the conclusion, draw some policy implications with your help. And canvas some reform options, probably without deciding on any specifically. Right, so the background. What is drug repositioning exactly? I'm going to be talking about developing new uses or what's known as medical indications for already authorised drugs. I don't really need to put already in there but I'm just emphasising this, they are already authorised. What I'm not talking about today is drug rescuing, when a drug was unsuccessfully developed for one use, one medical indication and then rescued and successfully developed for another. Why do we care about repositioning? In a nutshell, it's cheaper, faster and safer. Estimates on developing new compounds, they range 600 million to 2.5 billion US dollars. There's a lot of debate there, a lot of money. So repositioning costs about 25 to 30% of that price, so 70% cheaper. It's also safer because of course the drugs being used, we know a lot about it's safety profile. What is a new indication exactly? I want to spend a little bit of time here. And to understand this question, we must first appreciate that medical indications are often much more specific than just treating a condition or a disease. They include things, especially for cancers, they include things like age range, symptoms that people are suffering, line of treatment, that is this drug can only be used after 1, 2, 3 or 4 have failed or only as an adjunct. So they're very specific and they also include things like patient characteristics, obesity for example. This is included in the indication. What a new indication for the purposes of the legislation is any expansion of that. So for example, if a drug can only be used initially in anyone over the age of 35 and then it's repositioned for anyone over the age of 30, that is a new indication for the purposes of the legislation. Just as taking a drug that treated epilepsy to say treating pain is a new indication as well. Okay, the repositioning problem. So generally, we've covered this, I'll try and do it quite quickly. The prospect of selling drugs in competitive marketplace without IP is regarded as an insufficient because competitors can copy and then place their drugs on the market probably for a cheaper price without incurring all the development costs, a phenomenon commonly known as free writing. So the pharmaceutical industry is generally seen as the classic example for patent protection and for new compounds, it arguably works quite well, it probably works quite well. The situation is different for repositioning though. The claims that we'll see on specific repositioning, they're often quite weak, often what we call second medical use patterns. They have quite narrow protections, often limited to the use and they're often found invalid for a number of reasons. Then if they are found valid, there's often challenges to enforcement. A phenomenon commonly known as cross use can arise in where a generic version that hasn't been authorized for the new indication or at least on the label of the generic is still dispensed for the reposition indication. This happens due to the way the drugs are prescribed, dispensed and reimbursed. I won't go into the details of that. We know this phenomenon of cross use is common, but actually we don't know how common exactly. Right, so cross label use. So a case in point on this is of course just to illustrate is our favourite Warner, Lambert and generics. The decision concerned a second medical use pattern for the drug pregabalin. I'm going to go through this because I assume some people in this room aren't so familiar. The claims in the case were for treating neuropathic pain and pain in general. But the pattern was originally claimed for epilepsy. The UK Supreme Court invalidated the claims on the grounds of plausibility. The pattern showed that pregabalin was effectively treating inflammatory pain in mice. But the majority stated that the pain in mice was only implausibly linked to neuropathic pain or pain in general, so it was found invalid. The majority continued on to say that even if the pattern was valid, it was not infringed by the generic pregabalin when it was sold with a skinny label. And a skinny label being that it was only the label on the generic said it was only authorized for epilepsy. So they found that even the infringement couldn't be found against the generic in these circumstances because of the outward presentation of that generic drug. So people have argued and in particular Alistair Breckenridge, the former chair of the MHRA and Sir Robin Jacobs, the former Court of Appeal judge, do the cross-label use and these issues with invalidity is so significant that we should reform dispensing to solve the cross-label use problem and we should extend regulatory exclusivities to increase the incentive here. While we're talking about incentives, we need to appreciate though, what are the incentives in this space? Right, so I just said that patents were perhaps part of the problem in not incentivizing repositioning. At the same time, in another branch of the work we've been doing, we've actually found that patents are applied for second medical use patents are applied for at an increasing rate. I'll tell you who's been leading this work with us. So although there's concerns about enforcement and validity, they're applied for an increasing rate and increasing amount and they therefore must offer something. But other incentives and competitive advantages we're talking about are market demand, scientific lead time because the firm already knows, the pharmaceutical company already knows something about the drug, they have distribution networks, they have obliged drug surveillance due to the use they already have on market, relationships with regulators, relationships with reimbursement authorities, there's also likely to be some trade secrets perhaps in manufacturing and development and of course there's brand power that drug is on the market. The other incentive is of course what I want to focus on, the 8 plus 2 plus 1, 8 being data protection, 2 years of market protection and the plus 1. Let's look at that in a bit more detail. Regulation 7262004 article 1411 says that market protection for authorized drugs shall be extended by an extra year. If during the first 8 years of market protection, I need you to remember that, first 8 years of market protection, the marketing authorization holder obtains an authorization for one or more new therapeutic indications and you already investigated what a new therapeutic indication is, which bring significant clinical benefit in comparison with existing therapies that was covered by Trevor. Right, so now I'm just going to scoot through some historical notes. There was no incentive for repositioning prior to regulation 7262004 was confirmed in our under licensing authority in 1998. A 2000 EU commission report expressed concern about the lack of protection for significant new clinical indications. But at the same time the report admitted that there was very little hard evidence to support implementing an incentive for repositioning. Then, sorry a bit of a delay there, in the European Parliament Commissioner Lycanon argued that the plus 1 was the right balance between incentivizing innovators for new indications and allowing generic access, even though there wasn't much data. But we often make these decisions based on, well, talking to people on gut instincts perhaps. Procedural aspects of the plus 1, the plus 1 is only available for drugs applied for after 20 November 2005. I'm not talking about repositioning, I'm talking about the primary authorization when the drug was first authorized. Repositioning comes later. The drug authorization holder must apply for the plus 1 when they apply to reposition the drugs, including information on how they meet the significant clinical benefit test. The plus 1 can only be obtained once per drug. Right, so moving into the method now. The first stage of this method is that I downloaded the e-mail database of all the medicinal products authorized in the EU via the centralized procedure. This gave pan-EU rights, it began in 1995 and when I downloaded it mid last year consisted of 1,565 products. I then removed all the irrelevant entries such as generics, veterinary drugs and diagnostics, et cetera, leaving 742 drugs. In addition, the e-mail maintains a web-based European Public Access Report and EPAR, as it's commonly called, for each drug. And then within that EPAR they record all of the repositioning. The second stage of this method is where I split it up to answer those questions that I outlined at the start. So I divide the data into two groups. We have the drugs applied for before 20 November 2005 and then the after group drugs applied for after 20 November 2005. And then to answer this first question of whether it's increased the proportion, count the proportion that are repositioned at least once in the group. Importantly, repositioning only counts for the purposes of this analysis if it occurred within the first eight years of the primary authorization. To answer the second question, has it increased the number of times drugs are repositioned? I use the same grouping before and after as for the first question, but then count the average number of times a drug is repositioned, but only if it has been repositioned. Okay, now it comes to the crowd participation part that I promised everyone. Or as I like to think about it, avoiding hindsight bias part, because we all know things once we've been told the results. Okay, I'm going to ask you to predict the outcomes to two questions. And yes, before you say anything, a trap door will open below, unlike Matteo, Matteo is very generous. If you get the question wrong, I will open up the trap door. Please participate in this, it's a bit of fun. Okay, so I'd like a show of hands of whether, not the answers. And in fact, in your notes, don't turn the page, I've deliberately put a blank slide in there. Has the plus one increased the proportion of repositioned drugs? So I'll ask for a show of hands for increased, steady or decreased. So increased, raise your hand, steady, a few less, decreased. This is amazing crowd participation, thank you. Pass, no trap doors. Okay, and secondly, what proportion of drugs, this isn't actually my second research question, but it's important as you'll see later on, what proportion of drugs in my most recent sample, in the after sample, do you think were repositioned? And I'll ask this in increasing increments of 12.5%. It was just a little bit shorter, I was playing with time a bit here. Okay, so raise your hands if you think 0 to 12.5% of drugs are repositioned. 12.5 to 25%. That's quite a few. 25 to 37.5%. 37.5 to 50%. 50 to 62.5%. 62.5 to 75%. 75 plus. Very good, so far. No optimists in the room. Okay, the answers. Has the plus one increased, I should say, so largely people thought it would increase with, some people think it would remain steady, but a fewer number, very few people thought it would decrease. The largest amount was I think around 25, between 12 to 37.5%. And almost no one after that. Sorry though, I shouldn't have said that. Right, here are the answers. So has the plus one increased the proportion of repositioned drugs? In the before group, 256 primary drug authorizations, 108 of those were repositioned, giving a repositioned percentage of 42.2. In the after group, 125 primary drug authorizations, 56 repositioned, giving a percentage of 44.8. This is normally an increase of 2.6%, but do not pat yourselves on the back yet. Three extra drugs. I wouldn't write home about that one. It is not statistically significant using a chi-squared test, not even close. This suggests that what we're probably seeing there is just variation, particularly in a sort of small sample size. Another perspective on this though is of more fine-grained analysis. What I've done there is I've compared the most, the recent four years leading up to the commencement of the 2005, with the four years afterwards, which are actually the after group. And what we see there is that in the four years leading up to the commencement, 46.9% were repositioned, which is actually higher than afterwards, so maybe decreases right, and I don't think anyone really gets the pat on the back. But in reality, it hasn't changed. Steady, you guys win. Question two. Has the plus one increased the number of times drugs are repositioned? The before group, as we said before, 108 were repositioned corresponding to 215 events, which means on average, if it was repositioned, it was repositioned twice, or ratio two. A tiny bit of rounding there. In the after group, 56 drugs were repositioned corresponding to 103 repositioning events, giving it an average of 1.8. This is normally a decrease of 0.2 on average, or approximately nine less times drugs were repositioned. Not statistically significant, not even close. There's an interesting explanation here for drugs that we've already mentioned. The before group includes two drugs, Remicade and Humira, some of people's favourites, and were repositioned nine times, and another two that were repositioned eight times. The after group has no drugs that were repositioned six times or more, and what I think we're just seeing there is that random variations, sometimes drugs, were repositioned over time. Right, I should say Humira and Remicade, they were never eligible for the plus one. They're all in the before group. I initially applied for before. Policy implications. One implication from these results is that the plus one is an insufficient incentive, and I would have thought everyone in this room would just say, well, plus one, who cares? I think, yep, I generally agree with that. Sometimes I'm sure it's going to be very useful. But connected to this is the idea that other incentives or competitive advantages are generally more important. Market demand, first mover advantage, and Remicade in Humira. Humira was actually repositioned if we include all the time up to now, 23 times. Why do they keep doing that? They just expand the number of patients that can use it. We have first mover advantage. There's also patterns that operate in the background of this. We could sort of think of it as overlapping protection coming from the original molecule. SPCs, same sort of analysis, even though generally Trevor's completely right is not going to incentivize it sort of by itself. There might be new claims on formulations. First medical use claims, second medical use claims, although they have those difficulties we were talking about before. But there's also patterns that could be linked to drugs with diagnostics and drugs and delivery mechanisms. There's lots of other incentives at play here. Another possibility here is that we're already seeing quite high rates of repositioning. Most of you thought it was less than 45%, but it's been 45% for a very long time. Well, actually, my data doesn't go from 2011 up until now, but it's more challenged because I don't have that a year time frame to look at it, but I can tell you that it's sort of remained about the same. Another possibility here is that the cross-label use undermines the utility or the incentive effect of the plus one. That is very, very possible. We know that a cross-label use occurs. But how important is it exactly? As I said, we don't actually have much data on it at all. And we also know that there's various incentives or market factors to counteract the cross-label effect. Originator companies sometimes produce their own generic versions of the drug. Many drugs don't have generics. The FDA maintains a database of 400 market products, albeit many refer to the same compound, that do not have generics. They update it every month. Originator drugs continue to sell, particularly for biologics, and sometimes at quite high rates. And then we also see dosage of pharmaceutical forms, which sometimes prohibit or prevent in some way the cross-label use. High or low dosages for the reposition indication or pharmaceutical form change from cream to tablet or IV or used in a hospital used by a GP. Reform options. We could increase the plus one. It's probably not that much of an incentive. To plus three, like in the U.S. or more. What are the considerations here? The plus one didn't have effect. Why should plus three? How much more would we have to increase it by? And what cost? In my after data set, there were 14 drugs that obtained the plus one and another six that applied for it unsuccessfully. The data indicates that probably... Sorry, so we have 14 that got the plus one, but if we accept that the plus one didn't have an effect, then these 14 drugs that obtained the plus one probably would have done their repositioning afterwards. Sorry, without the benefit of the plus one. So we've just given a windfall to perhaps all of those 14, or maybe only 10 of them. So I think at the moment, my tentative conclusion that we should avoid increasing the plus one to plus three, unless Coge and Edison evidence says it will spur innovation that otherwise wouldn't occur. And importantly, and I don't know how to do this calculation, but it has to be cost-effective. Let's say in those 14, let's say four of them, only four were spurs. If the plus one didn't exist, that repositioning wouldn't have happened. But the other 10 would have happened regardless. Should we cross-subsidize those 10 with those four? Another option is to repeal the plus one. And I think this should be a definite option if we can work out how much the plus one costs. Is it 10 to hundreds of millions of euros or pounds? Or actually in those 14 drugs that we looked at, patent protection eclipses them all. And so actually it's not operating on the market. And so the plus one was just a backup option for those companies. Probably some research to be done there. We'd have to confirm that it isn't spurring in extra innovation. This research is not causative. It is observational. And then finally, what is the impact of cross-label use? We don't know. I would like to do some research on some people's judgments to instruct the NHS, not to prescribe. I think some interesting work could be done there. But we might do that in the future. I'd like to thank my co-authors, Mateo and Cappy, who we've already met. And the Nova Nordisk Foundation now. Funded. Thanks.