This episode is audio from the Empirical Evaluation of the Hatch-Waxman Act panel from the Engelberg Center's Hatch-Waxman at 40 and Beyond Symposium. It was recorded on September 26, 2024.
Scott Hemphill, Engelberg Center on Innovation Law & Policy, NYU School of Law
Bhaven Sampat, Arizona State University
Margaret K. Kyle, MINES ParisTech (Ecole des Mines) (commentator)
Announcer 0:00
Welcome to engelberg center live a collection of audio from events held by the engelberg center on innovation Law and Policy at NYU Law. This episode is audio from the empirical evaluation of the hatch Waxman act panel from the engelberg Center's hatch Waxman 40 and beyond symposium. It was recorded on September 26 2024 All
Scott Hemphill 0:29
right, why don't we get started? Hi, it's me again. I'm Scott Hemphill on the NYU faculty. My co author Bob and Sampat to my left, and I have been doing some work on the empirics of patch Waxman. We thought that 40 years in, and frankly, the presence of this conference today was a good convening moment to try to take stock of where we're at in terms of patenting, exclusivity, patent term restoration, we've, you know, had half an hour or so on narrative. Here's half an hour or so on on the numbers to try to trace out the growth in patenting on the orange book and what that has meant alongside paragraph four, challenges for the duration of effective exclusivity. And if this sounds like the kind of thing that Bob and I were working on about 10 years ago, that's right, this is basically the same kind of project updated to the present, adding an overlay of term restoration. So thinking about both arms of the hatch Waxman compromised both the pathway for generic entry on the one side and also the restoration. And there's some of the audience who very readily insert partial restoration, which will be a point of emphasis for us, partial restoration of the time spent in FDA trials and in review of the FDA so I'm not going to spend a lot of time on the key features of the hatchbacks map. You've already heard some of this. A key element here is this abbreviated new drug application process by which a generic can come to market without redoing the brand's clinical trials. Patents associated with the drug are listed on what most in the audience know is the orange book. It's for researchers, this really fun patent product linkage that is available in pharma and kind of hard to come by in other industries. So you can actually see the patent protection that the drug maker asserts is most pertinent, and they have an incentive for getting that right. And then some you have a pretty complete, not fully complete, but pretty complete set of the patent protection applicable to a drug. And of course, inclusion on the orange book has certain consequences for the existence and pace of generic entry, the 180 day exclusivity, about which you'll hear probably much more. Maybe we're lunch in the afternoon is a special bounty to the generic drug maker the first or they could be tied from first to challenge the patent on a brand name drug and not lose that litigation. They can win the litigation, or they could settle it and hold on to a special protection from other generics coming into the market. I'm going to mention one more the 30 month stay, which kicks in and blocks FDA approval when a brand challenges, sorry, when a brand takes up the challenge and engages in patent litigation with the generic so want to bring some data to this question, this question of, aren't patents going crazy? Doesn't that mean that branded protection is going to go really long? Or, on the other hand, is the rise of paragraph four challenges. That is patent challenges by generics eviscerate branded protection. The data that we're bringing to this question is the set of all non injectable new chemical entities. Those are the most innovative drugs, no active ingredient previously approved. That's the one that gives you the five years, or maybe it's four years, depending on how you count it, of exclusivity before generics can start trying to come in, we pulled together. Well, a group of us pulled together over time, sequential orange books in paper and in digital. The listing of NBR there, the National Bureau of Economic Research, is a reference to a database that's lodged there. There's an article that provides a kind of User's Guide that Lisa allet and myself and Bob and a couple of others, Heidi Williams, Mahindra salva, put together about a year ago. The patent challenges and generic approvals, as you'd expect, comes from the FDA. We have data on patent term restoration pulled together from. PTO and the Federal Register and the FDA, which is more of a thing than you would expect to pull have to pull together the patent characteristics. The key here, is it a what we describe as a primary patent, a patent on the active ingredient, or is it what we think of as a secondary patent? Kind of everything else that coding can be done a bunch of different ways. In this case, we're using IQ via data that does some of that patent typology, and the typology matches up quite closely with the hand collected work that we did 10 years ago. And then finally, drug sales from SSR health. So a couple of couple of main messages here, patents per drug, it's been going up. It's been going up pretty much throughout the life and hatchbacksman Act. Today, the count of patents is up to eight at the median, 10 at the 75th percentile. The mean is poking along at about six. The right hand picture here is just a count of patents excluding what we think of as late patents. So patents that were issued three or more years after the approval of the drug, you might think those are relatively less likely to bind against a generic who's trying to come in making the exact same thing as the brand as of the time of the brand's approval, we see a rise, if anything, more straightforward, rise, in patenting, if you exclude the late added ones, is there a secondary patent? So rather than just filing on the active ingredient, seeking a patent on one or more, let's say ancillary aspects of the drug, secondary patenting has always been with us. We're not going to be heard to claim that every secondary patent is invalid or not infringed, but I think it is. It is robust to say that all things equal the active ingredient patent is almost always a winner for the brand and the secondary patents by and large, or they're not always by and large or not. So we see a rise in secondary patents over time, which one might take as a kind of indicia of, I don't know, aggressiveness, some would say rapaciousness on the part of the brand, really pushing maybe out of necessity, maybe out of opportunity. It's hard to distinguish those depending on how you depending on how you look at it, raising the question, if you have more patents and more secondary patents, what does that mean for the pace of generic entry. Now this is what I'm about to show you. Now. Is not the right graph to look at for that question. This graph is a graph of nominal patent term by which we mean, as others have also measured the year of expiration of the last expiring patent right up and up and up and up and get to 18 years. At the top quartile, you can get north of 19 years, 20 years, 21 years. This is nominal term. This is what would happen if the generic never filed a paragraph for a challenge, never challenged the patent to the brand, with an exception I'm not going to get into here. It's the kind of hope for protection at some level. But you shouldn't take that as a measure of the actual protection, because, in fact, generic, generics can and do make an effort, ultimately successful effort, to to enter earlier. We see this in the form of these paragraph four challenges, which to repeat is a situation in which the generic says that applicable patent protection is either invalid or not infringed. We've long had since, let's call it the 1991 to 1993 approval cohort, a robust business in making these challenges and coming in early, we see this peaking, actually in the 2006 to 2008 cohort. One question, fairly invited by this, what happened in the last few cohorts? I have a question to this extra, this very expert audience, raise your hand during the Q and A and tell us what happened during these later, later cohorts? This is true for challenges overall. It's also true for early challenges, which you might think would help us work out whether this is just truncation, because it's not just truncation. Judging from the early challenges, we've not done a careful dive yet, I would say, to sort this out, it could be a change in the underlying composition of the drugs. Anyway, that's that's one thing that comes out of the data.
First challenges happen aggressively. Recall, this is a new chemical entity data set, so each of them got the five years of exclusivity, which for a paragraph, four challenge means, for some folks, is old hat. For some folks, this is just probably overwhelming jargon at the year four stage, after four years on the market, a paragraph four challenge by the generic can be launched that really tall bar that accounts for, I don't see it, not quite 75% of the data. That's the. N, C, E, plus four date. So one or more generics show up on the first day they possibly could to challenge it, right? And start that and start that clock ticking. They challenge one or more patents. That is not to say they challenge all the patents, and quite typically, they don't challenge all of them. So what are these patent challenges doing in aggregate? Well, they're pulling back on the extremes of nominal protection. And so you've got some drugs up here at the top of the distribution with 2530 35 years of nominal protection, but the effective market life is a lot smaller. You see our little average running here. It's like more in the more in the teens. This shows you the average by cohort. It's been poking along on average in this kind of 1112, 1312, 13, range. So one message here, I think, is this guy is not falling from the standpoint of repetitive, generic injury, that notwithstanding the rise in Orange Book patenting and not risk notwithstanding any number of shenanigans that I would be the first to decry, and we'll have some time to do that. I think in the afternoon. At a macro level, the amount of protection is relatively stable. We can break this out a little bit by sales quartile. So you see that as sales go up, the counter patents goes up, the nominal term goes up, the likelihood of a challenge goes up. The likelihood of an early challenge goes up. All of these things are, I think, fairly intuitive and interestingly, effective life, you know, is a little bit smaller than in the top part of the distribution. I'm going to leave out the which patents matter for now and turn it over to Bhaven.
Bhaven Sampat 12:02
I Scott, great. Thanks. So I'm going to talk about the other side of the ledger. The other side of the Grand Bargain was hatch wax given the patent term restoration. And as Scott mentioned, this is relatively new work at least, at least for us. And I'll do some basic descriptives. I'm going to talk about how the patent term restoration features of hatch Waxman interact with things like primary versus secondary patents, nominal versus effective term, and some other concepts that Scott raised. And then I'll say a little bit about implications for innovation and looking forward so the mechanics and mere economist and so there's people in the room who know this much, much better than me, but essentially, for new chemical entities, which is what we're looking at here, you can extend one patent per drug subject to there's various restrictions I won't go into, but there's caps, including there's a maximum amount of extended days, five years, two years for so called pipeline drugs that were in trials but not approved at the time hatch Blacksburg was was enacted, and also a total of a 14 year cap on the extended 14 year term on the total extended pap firms have a strategic choice then. And Becky, I think, has written about this, if not others, in terms of, you know, if you are faced with a choice and you have some primary you have a primary patent, typically on a drug, and secondary patents, which to extend the strong primary, or is it the secondary that lasts longer when faced with that choice, our data suggests that disproportionately you choose the primary patent. Firms choose the primary patent, north of 75% of the time they're choosing they're choosing the patent the primary This chart just shows the distribution of days of patent term extension. The left is all patent term extensions for the drugs in our sample. The right excludes the pipeline drugs that I just mentioned. And I think the thing that stands out here is just these big, these big masses at at the caps. So the two year caps, a lot of the drugs, including, like relatively recent drugs, are subject to the two year caps for pipeline drugs, and then the five year cap also seems to be important. That's that's sort of important, in a sense, patent term restoration is is incomplete, as Scott, as Scott mentioned, partly because of these caps. The other reason it's incomplete, and I didn't say this when I spoke about mechanics, is you only get back half of the time lost in clinical trials, plus essentially all the time lost in FDA review. So actually these caps, they end up binding for the majority of drugs in our set, 77 of the so of the of the 700 odd drugs that Scott spoke of, about 500 have at least one patent extended again, typically, the primary patent of those 77 are capped at five years of extension. 70 at two years. Of extension, and another 20, 26% the 14 year total cap applies. So if you mix and stir those, you just do some math in your head, the majority of drugs are subject to these caps on restoration. So together these caps, together with the partial nature of clinical trials, restoration, means that any restoration being offered here is, is incomplete. So what's the implication of that? And here I'm building, we're building on an extending work that Erica, Erica lightson has done in a really nice piece, distorted patents, which, which I'd recommend to you all. And the first, the top left graph, sort of built, builds on that. And it just shows you what's going on with the patent term for the extended patent relative to clinical trial length, which, which we get from the notices of final determination. And the bottom line is what it would look like without any patent term restoration. And the top line is what it looks like with patent term restoration, and you see two things. First, not surprisingly, with the term restoration, the line shifts up. So the patent term the patent term increases, but at the end of the day for the extended patent, the relationship remains negative. So it is the case that patents that are on drugs with long clinical trials have less term on the extended patent, even after the extension, here's where it gets, I think, more interesting and relates to some of the concepts that Scott introduced. What if we think about this at the drug level? What if we think about not just the extended patent term, but we roll it up to take into account all the patents at the drug level, look at what we call nominal patent term. If you look at nominal patent term versus clinical trial length, which is the top right chart, it flattens out. So these secondary patents are kind of flattening the whole thing. So that's an interesting part of the scope. But here's where the effective market life comes. So the effective market life, again, is the time to first generic entry on the bottom left. And if you look at that, that again, is negative. And what's going on here is kind of, I'm putting together several of the things I just I just mentioned and Scott did. First, firms typically extend the primary patents when given the choice. Second, you have other patents as well, but typically, especially for important drugs, especially for blockbuster drugs, those patents are challenged and generics prevail. So what's left standing is essentially the term that's sustained by the strong and extended primary patent, and at the end of the day, you continue to see this negative relationship between effective market life and clinical trial length. This creates distortions, as Erica and others have point pointed out. And I think is, is, is an issue. I mean, in the distortion, just to be clear, is that drugs that take longer to get through trials get less term. And that might be exactly, you know, the opposite of what we want from a from a policy perspective. And as I, as I read the legislative history of hatch Waxman, there actually wasn't a ton of attention to distortions of this sort, in a sense. I don't see where Al is, but there was no grand plan. He said, I mean, there was this sort of notion that we would extend patents, but not too much, right? And so you have these extensions for half the time in trials. Add on some caps. But what that means is that a fundamental distortion from an economic perspective, from a policy perspective, still remains with respect to with respect to the patents, and that's true both nominally and when you when you look at what happens at the end of the day, okay, so tying things up, then in terms of lessons and looking forward first, I think in policy analyzes, journalistic analyzes of these issues, the distinction between nominal patent term and effective market life are really, really important. I think we just have to be clear what we're talking about in these debates. There's a certain talking past one another. And that actually goes back to some of the deliberations around hatch Waxman in terms of what measure of patent term we're talking about. Second and Scott and I kind of gone back and forth on this curious to hear your thoughts. There is a sense in which paragraph four challenges they sort of work at ratcheting back these, these long nominal terms that are sustained or generated by secondary patents, but in the sense that you know the nominal is going up, but effective has been relatively stable in the decades since thatch waxa, but subject to all sorts of gaming. And we'll hear more about that, I think, later on in this conference, later on today, perhaps, and an extremely costly way to get back to square one, right? And so in this context, there are potential roles for perhaps the FDA, going beyond its ministerial role, to actually think more about what patents should be listed. And as Melissa and Michael have. Written about, sure, spoken about as well, more rigorous ex ante review by the US Patent and Trademark Office to improve upon the status quo. I'll say that there's an intermediate solution that is kind of a footnote, I think, in our 2011 piece, but I still really like it, which is
hatch Waxman pharmaceuticals. It's a unique context where we know what patents actually matter on a drug, where we typically don't by virtue of Orange Book listing. So what we you know, back then, there was this thing called RE examinations. They might still be on the books, but you don't hear so much about them anymore, but we argued for immediate reexamination of patents when when listed on the orange book, much like the second, you know, the PTO has done other things in other areas where there was patent quality concerns, so something like that, whether the RE exams or IPRS or whatever, might still be a useful thing to do. And then the last, the penultimate point I'll make, just on on restoration, is, I already said, if the Goal, goal is uniformity, restoration is sort of directionally correct, but incomplete. But I think we are not actually sure that the right policy goals should be uniformity. And we think that there is promise in looking to another feature of hatch Waxman, the exclusivity provisions, and building on remarks that Bob and I think al both made, like using these exclusivities to kind of, instead of, instead of patent terms, but not to get to uniformity, but maybe to tailor the amount of protection we give to the value of drugs, and maybe the, maybe the cost of drugs. And I think that's a useful thing to think about going forward, last but not least, just in terms of what we think would be useful going forward is, in some sense, hatch Waxman limits us in terms of how we think about things. There is this framework that's sort of naturalized by the act that we have one policy instrument, patents to do innovation, and then we have these good generics to come in and handle access and prices and affordability and competition, but that might be a limiting framework, and we see some of, I think, the costs associated with it and the games associated with it in the slides we've seen in the slides I just presented. But there's a number of other ways to do this, including policy ideas that were left on the cutting room floor in the decades that led up to hatch Waxman, direct government funding of clinical research trials, government generic production, tying patents to therapeutic efficacy, licensing, price controls. I'm not actually saying these are any of these are good or bad. I've got thoughts on that, but there's a range of other tools in our policy toolkit, beyond just patents and generic entry to handle these kind of balance, these sort of twin policy goals of innovation and access to promote public health. And I think that's the kind of conversation that would be useful to have going forward. So thanks, and I look forward to discussion, applause.
Margaret K. Kyle 23:09
Yes, someone will help put their slides. As that's being done, I'll just say thank you for the invitation to be here. I am based in Paris. But actually, for the year, I'm visiting NYU Stern, so it was not a long trip. Despite that, I will try and some of my comments to to bring a European perspective, since maybe that was the whole reason I was invited was to say something that, to add something to the discussion, to this audience. Okay?
Okay, well, maybe I'll Start without slides. I'm
okay, well, I was hoping to have a chart to show you to make this point. But as you, as you just heard with hatch Waxman, there are these competing goals, or an attempt to balance the incentives for innovation as well as create an environment where generic competition can take place. Overall, the big picture on innovation, we've seen essentially a fairly stable, effective exclusivity term over this period. What's kind of interesting is that while 20 years ago, about there was a big concern about a productivity crisis. And a worry that we weren't getting enough new drugs entering the market, et cetera. Since then, we've actually seen an increase in the number of new drugs brought to market. So we have stable, relatively stable, effective exclusivity, and we're seeing in the last couple of decades, in fact, a higher number of drugs brought to the market. So of course, we don't know exactly what the right level is, and different people can look at those trends and have different interpretations. So one interpretation might be, well, yeah, we're seeing a lot of me too drugs. So this is not valuable innovation, right? Others might say, well, despite the fact that there's that there's been this stable exclusivity, we've seen other big changes in other policies that will affect the incentives to innovate, like, let's say, an expansion of Medicaid, Medicare, Part D, expanding the market, the potential market, which is going to get more investment, etc. So it's, again, it's not clear exactly what the target number of enemies per year should be. I would say, though the good news is that it doesn't look like the fact that there's incomplete restoration of patent terms, it doesn't seem like that's having big negative effects on the overall level. I think there is a worthwhile discussion to have about the distortions introduced by the incomplete restoration, in particular, linked to the length of clinical trials. Another big picture point here. It is really unfortunate that I don't have the slides. If you look at how things play out in the US versus other countries. As we all know, prices on originator products are significantly higher in the US. What gets less attention is the fact that that generic markets are actually much better working in the US than in most other countries. So generic prices in the US are lower than they are in other developed markets. And hatch Waxman is one element, I would say, in making that happen. So it's not the only element. There's lots of other factors that create the incentives for generics to get into the market, but I think it's, it's worth pointing out. So here I'll try and and give you a little bit of information about how things work in another set of rich countries where there is no hatch Waxman, there's some points of commonality between Europe and the US, but some important differences. So in the absence of a hatch Waxman equivalent in Europe, how do what happens? So in Europe, there is longer statutory exclusivity on small molecule, molecule drugs in particular. So it's eight plus two plus one, eight years of data exclusivity, then an additional two years when generics have access to the data that can't enter the market, plus potentially one more year if the originator comes up with a new use, a new therapeutic use. So that is related to some of the points raised here at the end, and I think even in the first panel as well. If you don't use patents, if you use regulatory exclusivity, maybe you give yourself a little bit more flexibility in choosing the term linking it to therapeutic value. Europe doesn't go very far with that, but at least it's a step in that direction. There are also patent extensions through what are called supplementary protection certificates, which are also incomplete in the same way that patent extensions are through hatch Waxman. There's no patent linkage, so there's no orange book equivalent. The Orange Book generates huge benefits to all researchers in this area, because we have, thanks to that linkage, an ability for us to link to products, to patents, but there is no regulatory linkage in Europe. There's no equivalent of paragraph four, so there's no prize for challenging a patent. And what happens overall is you see far fewer secondary patents in Europe. Again, we can debate as to why that is, is it better screening by the EPO versus the USPTO? Is it just there's less incentive to build this wall as secondary patents, because there's less incentive for generics to challenge them, et cetera? There is a lower probability of generic entry. There is a longer wait for generics, and there's lower penetration of generics. So while there's a lot of problems that will get attention these, these couple of days about gaming of the hatch Waxman system, yeah, overall things actually on the generic side in the US, I would say work pretty well. Okay, I think there is a legitimate question as to how many of the hatch Waxman features are necessary to get us there. Also interesting, again, wish I had the chart to show you, but the effective protection period in Europe with a very different system, with more regulatory exclusivity, fewer patent battles. Turns out, in recent years, well, since about 2000 or so, 2005 To be approximately that of the US, so we don't necessarily need the litigation to claw back the time, right? You can set up a different system and get to approximately the same the same level. Now there's lots of other differences around policies, like mandatory substitution. Europe has only this month, established, and not for all of Europe. But there is finally a unitary patent. But before this patent was country by country, or protection was country by country, litigation was country by country, right? So the regulatory system for approving generics also is sort of in the process of evolving. But I think it's worthwhile thinking about some lessons for the US from another region, as well as lessons for Europe from the US experience. Maybe we can learn from each other's regulatory mistakes and successes on, for example, patent reform, so Europe tends to lean much more heavily on ex ante review, right post, Grant oppositions, et cetera. So we had some movement in that direction in the US over the last few years. So maybe that's encouraging the period of effective exclusivity. Do we want to determine that through patent litigation, or do we want to try and set it ex ante through these regulatory exclusivity periods? If you think that the patent litigation is leaving the best drugs with the highest profits and creating the best kind of innovation incentives, then maybe that's fine. But if you think the patent litigation is not necessarily being won by the best drugs, then there's more of a concern. Now we go through regulatory exclusivity terms, are regulators capable of assessing what is the therapeutic value? Can we do this in a way that's not also going to be gamed? So regulatory exclusivity periods have the advantage of starting start the clock once a drug reaches the market, but then maybe we're reducing the incentives to get through clinical trials as quickly as possible, to benefit patients as quickly as quickly as possible. So I think we do need to think carefully about about the potential for gaming those. And the last point I need, I will make, and again, I think this echoes other points, is that all of these policies interact with other policies and other institutions. So I don't think it would be the case that if Europe just adopted the hatch Waxman framework, that would necessarily see the same the same outcomes, because other institutions are different, and similarly, in the US, maybe it's not possible to just cherry pick what we think is working well and without thinking carefully about how this is going to interact with with other institutions and with other policies. And I'll end there and look forward to the discussion. Melissa, the question for you,
Speaker 1 32:48
Mars, super interesting. I was curious about what goes on in new uses in Europe, because I've always felt like because they tack on
Speaker 2 32:54
additional new use. Thank you for the regulatory exclusivity that you could argue they have more incentives to do new uses, and we tend to think in the US, because we have off label, right? And once the patents is no longer there and their generics right, that there's not as much of incentive as we may want to have all drugs for new uses. So I would just love if you could say a few words on that,
Margaret K. Kyle 33:20
so I don't have the breakdown of how many of the new drugs brought under a plus two also get the plus one. I think there's certainly a lot of attempt to get the plus one. I don't remember what, what fraction are successful in getting it in terms of
how many? So, I guess, the question would be, do we see drugs introduced in Europe that have no patent protection and therefore don't ever, don't even bother in the US market? But there are few examples, I think. But you know, what's always difficult in looking at innovation incentives is, it's a global market, so Europe overall is a pretty big market, but is it enough to move the needle on its own without a potential market in the US and a couple other places? You know? Again, there are a few examples I can think of, but it's not it's not the case that Europe is getting a huge number of old drugs for new uses that that we're not aware of in the US? Bruce, thanks.
Scott Hemphill 34:28
We can hear you, but still,
Bruce Wexler 34:30
yeah, I just wanted to address Scott, your question about the decline, and I have it's based on anecdotal experience managing a large IP practice that's representing a wide swath of small, medium and large drug companies, also looking out at when drug challenges are going to happen, because that's when we would expect to see filings and litigation. So I remember there was a period of decline. It was noticeable, and it coincided, I think, around the time you were talking about and we saw two things. Happening at that time. One, there was a large swath of generics who were popping up names we hadn't seen before. They would just come and seem to be coming out of the woodworks, and they were all as like a feeding frenzy of a lot of generics on this, on this, on the other side of it, there was, we weren't seeing as many sort of NC minus one dates, and so there wasn't as much sort of food available for the generics. And those two things combined to make for a very challenging environment. I think for lawyers, what I would say is, because you didn't have litigation, so I just, I don't know if that's going to coincide with your research, but that's anecdotal. Thank
Scott Hemphill 35:40
you, Artie, so
Arti Rai 35:43
outside the wonderful world of biopharma, there's been a lot of interest in patent term adjustment, which can interact in interesting ways with patent term extension. Are you looking at patent term adjustment as well?
Bhaven Sampat 35:58
So not explicitly. I mean, the adjustments are built into the orange book expiration dates that we're looking at. I mean, we we have the data. But so what would you suggest would be interesting to do with us in this context? Yeah, I
Arti Rai 36:16
would just be curious about whether you know it. So whether adjustment is playing any kind of role in any of this at the end of the day? I mean, I don't think so, but it would just be a good check on your data. Yeah, for sure.
Unknown Speaker 36:31
Let's talk some more. I
Scott Hemphill 36:33
wonder if there's some subset of cases where it changes the election that's hard to do all the thinking on that. Which one? Because it's right now we think Street Primary but I wonder there's something going on there under the surface. If there was something jagged about the bonus they're already getting from PTA, that would then change the PTE selection in some way that maybe we could do a finer grained thinking about, yeah, super
Speaker 3 37:02
interesting to hear Scott and Bob and a little bit more about how product hopping might fit into your framework. So you talk about time to first generic entry, but I'm wondering if what we care about is not just generic entry, but effective generic competition. And is that accounted for in your data? If not, have you thought about how you might look at it? No. And I
Scott Hemphill 37:21
think it's a little worse than not counted in the data. It's also not counted in the like, the basic presentation, the kind of like, I mean, I'm a little uncomfortable in the persona of sanguine about all these different strategies. And like, we do have this big picture result that effective term isn't crazy out of whack. And so however much reverse payment there is, there isn't enough to like, or it's not effective enough to make some drastic change across, you know, hundreds and hundreds of drugs. But that is of not just small comfort, but zero comfort when it comes to product offering, because they're switching to it. I mean, typical, right? We switch into a different NDA, and so we wouldn't observe that at all. Now, in the paper, we talked about this sum. We talked about product hopping the book, not in a way that is informed by the data, I think, any any thoughts about so we're open for business, I should say,
Speaker 4 38:18
awesome. So like, we have a first draft of a paper, we're looking
Scott Hemphill 38:24
for help. On the fallen in paragraph four is, we're also looking for help. I know somebody in this audience is gonna know the answer to this. Like, we see fewer extensions than we would expect, fewer even attempts to extend. Does that mean we have the category wrong? The denominator is different from what we think, or is it like, Oh, you don't know, Scott, about the thus and such reason. So that's another to do for everybody. I bet Erica knows if there's, if there's a data exercise that would share light on this that you know, having a big old set of NCS and spending time thinking about the patents like we were, like, very interested to
Margaret K. Kyle 38:59
hear it. Can I just add something about the product topping. If the rest of the healthcare market was working efficiently, this really wouldn't be a problem, right? Because payers and patients should be able to see through the fact that this is just repackaging of something old, if that's all it is, right? If there really is nothing new in this hopped product, it's the sort of strategy that, yeah. I mean, maybe it's, it's possible to gain some aspect of FDA approval and hatch Waxman, but it's only possible because we have so many other problems in healthcare markets with information and incentives, et cetera. And somehow, I mean, I just think as a general rule, we should be focusing on fixing a lot of those problems first, because they would have effects, not just on this particular the product copying issue, but you know, in general, there's just a huge gain to to that. Yes,
Speaker 5 39:56
yeah, kind of building, building on Rachel's question. I. It looked like you were looking at drug approval and not actual market entry. Even with market entry, maybe a better measure would be something like what Michelle Melo did with reduction in market share when the generic comes on.
Scott Hemphill 40:14
Yeah, go ahead.
Speaker 5 40:15
So the second question was verbatim, which is the clinical trial kind of extensions. Is there some sort of selection bias going on? Like, is it that the trials that take longer are for drugs that are, you know, arguably could be maybe worse or not, you know? Like, it would be interesting to kind of see which drugs are actually going through the longer approval cycle, maybe that's why everything kind of flattens out. Well,
Bhaven Sampat 40:45
I don't know if I call it a selection bias, but it is true. So implicitly in how I presented it, there is the idea that there is a distortion, and that distortion is bad because drugs that take longer to get through trials because they're harder. Or, you know, the work that Heidi Williams and Buddhism Roy have done is that, you know, and that's the sort of distortion from a bad perspective. It could be that, you know, the 13th, the me too, drugs, right? Like, you know that those take longer, and so we don't actually know we're Yeah, and I think that's yeah. That's something we should clarify in the pros. We don't actually know if those long trials are better or worse. Yeah, the economics literature so far is focused on this being a distortion. But whatever it is, I mean, it seems like it's arbitrary, and just relate. It's not. It's not by design. And I think we should think a little bit more by design about what we want in terms of different types of exclusivity for different
Scott Hemphill 41:46
types of drugs. And on your first point, I totally hear you on approval. You read that correctly. It's approval that we're using. I think I agree with you that, like some kind of actual entry moment, there's different ways of doing that we you know, in our earlier work, we've done it with press releases. You can do it with a share shift. If you have the right it's another field you got to bring in right? It's additional data. What we found in the past is that the correlation is just so high that it doesn't change qualitative results, though it could if the if the gap was, if the gap was big enough, and we might be losing some precision, and it's not like these are, it's not a zero mean thing, I think between the two. So I think that's something we're we're considering whether to make the additional investment to also, you know, nail that down. And for certain policy questions, I think would be quite important to potentially tease that out. Yeah, Bob,
Bob Armitage 42:51
just an observation in terms of longer clinical trial results, my experiences, unprecedented therapies, breakthrough drugs, chronic drugs, where the clinical trials naturally last longer. For example, Alzheimer's medicines, clinical trials have to last long enough years sometimes. So there really is a strong correlation. But looking at all this data, it seemed to me that we've created this system designed to waste huge amounts of money. We have a very complex patent extension law. It actually takes now years for the FDA and the patent office to process an extension. Usually you do multiple patents and then pick the correct one later. And then the biggest waste, just if I look at this big spike, is the NC minus one four year data. Your data suggests people wait. Most generics Wait, say, almost a dozen years to get on the market, and yet, they do all this early work before the product really, they have to start that work before you even know if the product's going to be successful in the marketplace. So my question is, isn't the solution here simply to say, look, if you list a patent in an orange book, you get your 14 years. It's just done. The application's one sentence long. Patent listed in Orange Book, give me 14 years, and then you move the four year period to 10 years. And then the only patents that get challenged are the ones that appear, statistically to be easy to overcome, so you're either not sued and there's no litigation, or you have a high probability of just knocking them out at relatively low cost, and that ends this perversity of patent term being less for drugs that are more innovative. Is that really where we ought to be headed?
Scott Hemphill 44:40
I mean, we mean we have five responses across the three of us. I mean speaking for myself, I'm I'm sympathetic to a simpler system, and I'm sympathetic to a simpler system, one, because our current system is extremely expensive, and two, because you. There's even if our average this didn't really come out as well as it could have in my initial presentation, like even if the average came out in some place. Let's imagine we all agreed that, like, 12.7 is exactly right. The variation around that has no plausible relationship to other measures of interest. And we can fight about what the other measures of interest are, but measures of interest are, but there's no reason to think. Let me pick one like take reverse payments for just a minute. We'll talk about settlements in a while. Reverse payments tend to be in settings, disproportionately where the brand is going to lose the patent litigation. That's where the incentive is greatest to spend the amount of money, the the the the benefit, the incremental benefit, from an exclusivity standpoint, is most attractive in places where there's the biggest gap at least measured from patent. And so if you think patent is in some rough way, measuring something useful, it's weird that we would seek recourse to this instrument to the greatest extent in the set of circumstances where the patent is is weakest. This is one among lots of examples in which there's no reason for us to think that strategies that are patent extending, that are exclusivity extending, are well matched to the set of therapies that we think we might want to have a longer duration on, and so a simpler system that bleached out that basically random or even harmful complexity, I think would be valuable.
Bhaven Sampat 46:36
So I get to give reasons two, three and four. No quick. I think there is selective lengthening and selective reduction that doesn't correspond to anything clinically or economically important. And so in some sense, a move towards uniformity, along the lines you suggest a one and done type system does make sense, as I alluded to at the end of my remarks. My own preference would be to move away from uniformity and try to tailor the extent of the reward a little bit more to the board a little bit more to the therapeutic contribution through exclusivity.
Margaret K. Kyle 47:07
But so I also would be for figuring out a way to tie exclusivity to therapeutic benefits. But again, I'll come back to the potential for gaming. So if, if a company has two drug candidates for the same for the same disease, they're roughly the seeing the same success in clinical trials. If they know that no matter what they get 14 years from the date they enter the market, they might be tempted to do one, wait the 14 years then launch the other, which is not the case under the patent system, right? Because with the patent you're you're not going to get that exclusivity, given that the clock starts at the point of development. So if so you know, so if I have two drug candidates, and I know that I'm going to get 14 years of exclusivity on each one of them, then I might be tempted to stagger them so that I get 14 and then I launch my new one and get another 14 which and that's something that would not be possible if they both have patents that start around the same time. So it's this kind of gaming that I would, I would worry a bit
Scott Hemphill 48:08
about, and that view is conditional on the other patent existing, or where you feel a severe risk that someone else is going to patent it if you don't,
Margaret K. Kyle 48:17
right, so right, which is probably true sometimes, but like, if
Scott Hemphill 48:21
somebody is just shelving, you know, the next generation of, let's say, some HIV therapy and weights to, you know, to file, and nobody else files like you could imagine.
Margaret K. Kyle 48:31
So maybe there's enough offsetting, offsetting controls for that. But that's,
Scott Hemphill 48:36
is that a hook over there? Or do we have time for another Yeah, five people are reacting, and now let's go check all I
Bob Armitage 48:41
was at Lilly, I was once called our socialist General Counsel, but I'm actually a capitalist, and what happens is competition prevents that from happening. In other words, in most cases, you have fast followers coming along with improved therapies. You look at almost any area, and nobody's sitting on technology in any area, including drugs. And the second thing is, isn't there a marketplace that's going to determine value, or should determine value? In other words, I put a drug on the market, I do the work I have. If I it's an innovative mouthwash, I get the full patent term. It's typically 20 years. But if it doesn't taste good, doesn't have any benefits, it just doesn't get a market. And I think we need to have a marketplace that's truly capitalistic, that drugs benefit. Drugs get high value, drugs get high reward. Otherwise they don't. Okay.
Margaret K. Kyle 49:39
So again, this comes back to the product hopping kind of argument as well. I agree that if, if markets are working well, and we have good information about the true value of every product, then that's going to be reflected in the price and the market share. And you know, these patent terms matter much less or the potential for abuse like product hopping. Is just much reduced because everybody sees through it. But that's if that's not the case, and that's a concern for a lot of people, that that competition is not working as effectively. Then let
Scott Hemphill 50:15
me get Nicholson
Speaker 6 50:19
in here. It seems super weird that in totally different legal regimes, we have approximately the same period of effective exclusivity. And it also seems super weird that in the hurling of tons of money at lawyers to develop a really fun, exciting shenanigans all over the place in this space, the answer is, the overall time doesn't seem to change. Like, what's up with that? Why is this a constant of the order? Why does this seem to be a constant of the field? Like that? I would not have predicted that at all. What's going on?
Scott Hemphill 50:53
Well, I don't really know. I mean, I think like, but it could be coincidence. These commanders are at the same place. There is the fact that it's, I mean, it's a possibility result of it, right? I mean, I think you present it as a kind of possibility result that you could get to about the same place with a different system, you know? I mean, the average is probably disguising some stuff, right? Also, I think we have reason to be worried, even if we're talking about 10 month end of life heroic efforts to eke out another six or $7 billion on a small molecule. Or it's, you know, or it and its close cousin, when we think about product hopping, we got a lot of these where, like, the claimed delay is like 18 months and or 20 months. It's rare that it's like six or seven years, but a couple of those way back in the 90s. But today, like this, this is being this ground game is being fought for inches that are worth, you know, $100 million or so an inch. And if it's happening across a couple dozen drugs, you know, that might get washed out in a cohort of 150 or something, right? So it could be that our our macro sanguineness, you know, bleaches out some of some real action. I mean, these aren't sales weight. Or we could do this in the sales way too, and sort of see the quantiles pick up some of that, right? So we're not, we're not seeing, we're not seeing, at the top quartile, some boost in effect of exclusivity, right? So even when you cut it a little bit finer, you're not seeing, you're not seeing some, some big gain. I think
Bhaven Sampat 52:38
it's a great question. I don't know much about Europe, but there is, in some context, interdependence of the policy debates. And in ways I talked about the naturalization. I mean, there's a sense in which, you know, 12 years sustained by primary patents in the US becomes the kind of baseline for thinking about how you want to meet policy interventions in other contexts, in the US, but also globally. And that, that's, that's a super interesting question.
Announcer 53:00
Now I'm going to give you the hook. You gave me the hook. Yep, right the second.
The engelberg center live podcast is a production of the engelberg center on innovation Law and Policy at NYU Law, and is released under a Creative Commons, Attribution, 4.0 International license. Our theme music is by Jessica Batke and is licensed under a Creative Commons, Attribution, 4.0 International license.