About this Author
College chemistry, 1983
The 2002 Model
After 10 years of blogging. . .
Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases.
To contact Derek email him directly: firstname.lastname@example.org
August 31, 2012
Eli Lilly has been getting shelled with bad news recently. There was the not-that-encouraging-at-all failure of its Alzheimer's antibody solanezumab to meet any of its clinical endpoints. But that's the good news, since that (at least according to the company) it showed some signs of something in some patients.
We can't say that about pomaglumetad methionil (LY2140023), their metabotropic glutamate receptor ligand for schizophrenia, which is being halted. The first large trial of the compound failed to meet its endpoint, and an interim analysis showed that the drug was unlikely to have a chance of making its endpoints in the second trial. It will now disappear, as will the money spent on it so far. (The first drug project I ever worked on was a backup for an antipsychotic with a novel mechanism, which also failed to do a damned thing in the clinic, and which experience perhaps gave me some of the ideas I have now about drug research).
This compound is an oral prodrug of LY404039, which has a rather unusual structure. The New York Times did a story about the drug's development a few years ago, which honestly makes rather sad reading in light of the current news. It was once thought to have great promise. Note the cynical statement in that last link about how it really doesn't matter if the compound works or not - but you know what? It did matter in the end. This was the first compound of its type, an attempt at a real innovation through a new mechanism to treat mental illness, just the sort of thing that some people will tell you that the drug industry never gets around to doing.
And just to round things off, Lilly announced the results of a head-to-head trial of its anticoagulant drug Effient versus (now generic) Plavix in acute coronary syndrome. This is the sort of trial that critics of the drug industry keep saying never gets run, by the way. But this one was, because Plavix is the thing to beat in that field - and Effient didn't beat it, although there might have been an edge in long-term followup.
Anticoagulants are a tough field - there are a lot of patients, a lot of money to be made, and a lot of room (in theory) for improvement over the existing agents. But just beating heparin is hard enough, without the additional challenge of beating cheap Plavix. It's a large enough patient population, though, that more than one drug is needed because of different responses.
There have been a lot of critics of Lilly's research strategy over the years, and a lot of shareholders have been (and are) yelling for the CEO's head. But from where I sit, it looks like the company has been taking a lot of good shots. They've had a big push in Alzheimer's, for example. Their gamma-secretase inhibitor, which failed in terrible fashion, was a first of its kind. Someone had to be the first to try this mechanism out; it's been a goal of Alzheimer's research for over twenty years now. Solanezumab was a tougher call, given the difficulties that Elan (and Wyeth/Pfizer, J&J, and so on) have had with that approach over the years. But immunology is a black box, different antibodies do different things in different people, and Lilly's not the only company trying the same thing. And they've been doggedly pursuing beta-secretase as well. These, like them or not, are still some of the best ideas that anyone has for Alzheimer's therapy. And any kind of win in that area would be a huge event - I think that Lilly deserves credit for having the nerve to go after such a tough area, because I can tell you that I've been avoiding it ever since I worked on it in the 1990s.
But what would I have spent the money on instead? It's not like there are any low-risk ideas crowding each other for attention. Lilly's portfolio is not a crazy or stupid one - it's not all wild ideas, but it's not all full of attempts to play it safe, either. It looks like the sort of thing any big (and highly competent) drug research organization could have ended up with. The odds are still very much against any drug making it through the clinic, which means that having three (or four, or five) in a row go bad on you is not an unusual event at all. Just a horribly unprofitable one.
+ TrackBacks (0) | Category: Cardiovascular Disease | Clinical Trials | Drug Development | Drug Industry History | The Central Nervous System
August 30, 2012
Here's yet another chance to play the human biology game that might as well be called "Now what?" That's when we find that what we thought we knew is actually wrong, more complicated, or a sign of something else entirely.
Today's entry is niacin. As many readers know, it looks like it should be a promising therapy for patients whose lipoproteins are out of whack. It lowers LDL, raises HDL, lowers free fatty acids, and lowers triglycerides, and all those things are supposed to be good. (As came up in the comments yesterday's post, though, the evidence is pretty strong for that first proposition, but not as solid for the others). Still, if you went around to thousands of cardiologists and asked them if they'd be interested in a therapy that did those four things, you'd get a resounding "Yes".
So why hasn't niacin taken over the world? Because of the side effects. It has to be taken in rather stiff doses to show the lipid effects, and those tend to cause a nasty skin flush reaction, which is apparently unpleasant enough that most people won't put up with it. Various attempts have been made to abrogate this, with the most direct assault being Merck's (failed) Cordaptive.
The flushing is thought to be mediated through the receptor GPR109A, via a prostaglandin pathway. Unfortunately, it's also believed that niacin's beneficial effects are mediated through that receptor, too, via some mechanism that starts with the lowering of free fatty acids. If you knock out the receptor in mice, you get no skin flushing, but no FFA lowering, either.
We must now revise that idea. A new paper tests that hypothesis with two non-niacin agonists, MK-1903 (a compound via Arena Pharmaceuticals, I believe) and SCH900271, and their effects in humans. They also report niacin's effects in the receptor knockout mice, claiming that although the FFA lowering does indeed disappear, that the downstream lipid effects remain. (That surprises me; I'd thought that had already been studied).
But the human data are especially revealing. The two new agonists do indeed show FFA effects, as you'd expect from compounds hitting GPR109A. But they do not show chronic free fatty acid lowering, nor do they have the desired downstream effects on blood lipids. So it appears inescapable that niacin's effects are going through some other pathway, one that doesn't depends on GPR109A or its (transient) free fatty acid lowering. Back to the drawing board everyone gets to go.
But niacin has been heading there already. Readers may remember a trial of a niacin-and-statin combination had to be stopped early because the cardiovascular effects were (alarmingly) going in the other direction. Not only was there no benefit, but there seemed to be active harm. Taken together, all this tells us that we have an awful lot to learn about some things that we thought we were starting to understand. . .
+ TrackBacks (0) | Category: Cardiovascular Disease
August 29, 2012
Nature is out today with a paper on the results of a calorie-restriction study that began in 1987. This one took place with rhesus monkeys at the National Institute of Aging, and I'll skip right to the big result: no increase in life span.
That's in contrast to a study from 2009 (also in rhesus) that did see an extension - but as this New York Times article details, there are a number of differences between the two studies that confound interpretation. For one thing, a number of monkeys that died in the Wisconsin study were not included in the results, since it was determined that they did not die of age-related causes. The chow mixtures were slightly different, as were the monkeys' genetic background. And a big difference is that the Wisconsin control animals were fed ad libitum, while the NIA animal were controlled to a "normal" level of calorie intake (and were smaller than the Wisconsin controls in the end).
Taken together with this study in mice, which found great variation in response to caloric restriction depending on the strain of mouse used, it seems clear that this is not one of those simple stories. It also complicates a great deal the attempts to link the effect of various small molecules to putative caloric restriction pathways. I used to think that caloric restriction was the bedrock result of the whole aging-and-lifespan research world - so now what? More complications, is what. Some organisms, under some conditions, do seem to show longevity effects. But unraveling what's going on is just getting trickier and trickier as time goes on.
I wanted to take a moment as well to highlight something that caught my eye in the Times article linked above. Here:
. . .Lab test results showed lower levels of cholesterol and blood sugar in the male monkeys that started eating 30 percent fewer calories in old age, but not in the females. Males and females that started dieting when they were old had lower levels of triglycerides, which are linked to heart disease risk. Monkeys put on the diet when they were young or middle-aged did not get the same benefits, though they had less cancer. But the bottom line was that the monkeys that ate less did not live any longer than those that ate normally. . .
Note that line about "benefits". The problem is, as far as I can see (Nature's site is down as I write), the two groups of monkeys appear to have shown the same broad trends in cardiovascular disease. And cardiovascular outcomes are supposed to be the benefits of better triglyceride numbers, aren't they? You don't just lower them to lower them, you lower them to see better health. More on this as I get a chance to see the whole paper. . .
+ TrackBacks (0) | Category: Aging and Lifespan | Cardiovascular Disease | Diabetes and Obesity
Here you go, from IKA. If you can make it up to about 1:52 or so, that's when the traditional hard-sell starts. But up until then, it's pretty painful, and not least because the model playing a chemist is evaporating a bright green solution (sure thing) and the receiving flask is light blue (oh yeah). More unlikely colors are to be seen in the sales-pitch part of the video that follows, though, but at least there's no acting, or whatever that's supposed to be. Yikes.
+ TrackBacks (0) | Category: Chemical News
Startup biopharma companies: they've gotta raise money, right? And the more money, the better, right? Not so right, according to this post by venture capitalist Bruce Booth. Companies need money, for sure, but above a certain threshold there's no correlation with success, either for the company's research portfolio or its early stage investors. (I might add that the same holds true for larger drug companies as well, for somewhat different reasons. Perhaps Pfizer's strategy over the last twenty years has had one (and maybe only one) net positive effect: it's proven that you cannot humungous your way to success in this business. And yes, since you ask, that's the last time I plan to use "humungous" as a verb for a while).
There's also a fascinating look back at FierceBiotech's 2007 "Top Deals", to see what became of the ten largest financing rounds on the list. Some of them have worked out, and some of them most definitely haven't: 4 of the ten were near-total losses. One's around break-even, two are "works in progress" but could come through, and three have provided at least 2x returns. (Read his post to attach names to these!) And as Booth shows, that's pretty much what you'd expect from the distribution over the entire biotech industry, including all the wild-eyed stuff and the riskiest small fry. Going with the biggest, most lucratively financed companies bought you, in this case, no extra security at all.
A note about those returns: one of the winners on the list is described as having paid out "modest 2x returns" to the investors. That's the sort of quote that inspires outrage among the clueless, because (of course) a 100% profit is rather above the market returns for the last five years. But the risk/reward ratio has not been repealed. You could have gotten those market returns by doing nothing, just by parking the cash in a couple of index funds and sitting back. Investing in startup companies requires a lot more work, because you're taking on a lot more risk.
It was not clear which of those ten big deals in 2007 would pay out, to put it mildly. In fact, if you take Booth's figures so far, an equal investment in each of the top seven companies on the list in 2007 would leave you looking at a slight net loss to date, and that includes one company that would have paid you back at about 3x to 4x. Number eight was the big winner on the list (5x, if you got out at the perfect peak, and good luck with that), and number 9 is the 2x return (while #10 is ongoing, but a likely loss). As any venture investor knows, you're looking at a significant risk of losing your entire investment whenever you back a startup, so you'd better (a) back more than one and (b) do an awful lot of thinking about which ones those are. This is a job for the deeply pocketed.
And when you think about it, a very similar situation obtains inside a given drug company. The big difference is that you don't have the option of not playing the game - something always has to be done. There are always projects going, some of which look more promising than others, some of which will cost more to prosecute than others, and some of which are aimed at different markets than others. You might be in a situation where there are several that look like they could be taken on, but your development organization can't handle so many. What to do? Partner something, park something that can wait (if anything can)?Or you might have the reverse problem, of not enough programs that look like they might work. Do you push the best of a bad lot forward and hope for the best? If not, do you still pay your development people even if they have nothing to develop right now, in the hopes that they soon will?
Which of these clinical programs of yours have the most risk? The biggest potential? Have you balanced those properly? You're sure to lose your entire investment on the majority - the great majority - of them, so choose as wisely as you can. The ones that make it through are going to have to pay for all the others, because if they don't, everyone's out of a job.
This whole process, of accumulating capital and risking it on new ventures, is important enough that we've named an entire economic system for it. It's a high-wire act. Too cautious, and you might not keep up enough to survive. Too risky, and you could lose too much. They do focus one's attention, such prospects, and the thought that other companies are out there trying to get a step on you helps keep you moving, too. It's not a pretty system, but it isn't supposed to be. It's supposed to work.
+ TrackBacks (0) | Category: Business and Markets | Drug Development | Drug Industry History
August 28, 2012
There's an odd retraction in the synthetic chemistry literature. A synthesis of the lundurine alkaloid core from the Martin group at Texas was published last year, and its centerpiece was a double-ring-closing olefin metathesis reaction. (Coincidentally, that reaction was one of the "Black Swan" examples in the paper I blogged about yesterday - the initial reports of it from the 1960s weren't appreciated by the synthetic organic community for many years).
Now the notice says that the paper is being retracted because that RCM reaction is "not reproducible". (The cynical among you will already be wondering when that became a criterion for retraction in the literature - if it works once, it's in, right?)
There are more details at The Heterocyclist, a blog by the well-known synthetic chemist Will Pearson that I've been remiss in not highlighting before now. While you're there, fans of the sorts of chemicals I write about in "Things I Won't Work With" might enjoy this post on the high explosive RDX, and the Michigan chemist (Werner Bachmann) who figured out how to synthesize it on scale during World War II.
+ TrackBacks (0) | Category: Chemical News | The Scientific Literature
August 27, 2012
What's a Black Swan Event in chemistry? Longtime industrial chemist Bill Nugent has a very interesting article in Angewandte Chemie with that theme, and it's well worth a look. He details several examples of things that all organic chemists thought they knew that turned out not to be so, and traces the counterexamples back to their first appearances in the literature. For example, the idea that gold (and gold complexes) were uninteresting catalysts:
I completed my graduate studies with Prof. Jay Kochi at Indiana University in 1976. Although research for my thesis focused on organomercury chemistry, there was an active program on organogold chemistry, and our perspective was typical for its time. Gold was regarded as a lethargic and overweight version of catalytically interesting copper. More- over, in the presence of water, gold(I) complexes have a nasty tendency to disproportionate to gold(III) and colloidal gold(0). Gold, it was thought, could provide insight into the workings of copper catalysis but was simply too inert to serve as a useful catalyst itself. Yet, during the decade after I completed my Ph.D. in 1976 there were tantalizing hints in the literature that this was not the case.
One of these was a high-temperature rearrangement reported in 1976, and there was a 1983 report on gold-catalyzed oxidation of sulfides to sulfoxides. Neither of these got much attention, as the Nugent's own chart of the literature on the subject shows. (I don't pay much attention when someone oxidizes a sulfide, myself). Apparently, though, a few people had reason to know that something was going on:
However, analytical chemists in the gold-mining industry have long harnessed the ability of gold to catalyze the oxidation of certain organic dyes as a means of assaying ore samples. At least one of these reports actually predates the (1983) Natile publication. Significantly, it could be shown that other precious metals do not catalyze the same reactions, the assays are specific for gold. It is safe to say that the synthetic community was not familiar with this report.
I'll bet not. It wasn't until 1998 that a paper appeared that really got people interested, and you can see the effect on that chart. Nugent has a number of other similar examples of chemistry that appeared years before its potential was recognized. Pd-catalyzed C-N bond formation, monodentate asymmetric hydrogenation catalysts, the use of olefin metathesis in organic synthesis, non-aqueous enzyme chemistry, and many others.
So where do the black swans come into all this? Those familiar with Nasim Taleb's book
will recognize the reference.
The phrase “Black Swan event” comes from the writings of the statistician and philosopher Nassim Nicholas Taleb. The term derives from a Latin metaphor that for many centuries simply meant something that does not exist. But also implicit in the phrase is the vulnerability of any system of thought to conflicting data. The phrase's underlying logic could be undone by the observation of a single black swan.
In 1697, the Dutch explorer Willem de Vlamingh discovered black swans on the Swan River in Western Australia. Not surprisingly, the phrase underwent a metamorphosis and came to mean a perceived impossibility that might later be disproven. It is in this sense that Taleb employs it. In his view: “What we call here a Black Swan (and capitalize it) is an event with the following three attributes. First, it is an outlier, as it lies outside the realm of regular expectations, because nothing in the past can convincingly point to its possibility. Second, it carries an extreme impact. Third, in spite of its outlier status, human nature makes us concoct an explanation for its occurrence after the fact, making it explainable and predictable.”
Taleb has documented this last point about human nature through historical and psychological evidence. His ideas remain controversial but seem to make a great deal of sense when one attempts to understand the lengthy interludes between the literature antecedents and the disruptive breakthroughs shown. . .At the very least, his ideas represent a heads up as to how we read and mentally process the chemical literature.
I have no doubt that unwarranted assumptions persist in the conventional wisdom of organic synthesis. (Indeed, to believe otherwise would suggest that disruptive break- throughs will no longer occur in the future.) The goal, it would seem, is to recognize such assumptions for what they are and to minimize the time lag between the appearance of Black Swans and the breakthroughs that follow.
One difference between Nugent's examples and Taleb's is the "extreme impact" part. I think that Taleb has in mind events in the financial industry like the real estate collapse of 2007-2008 (recommended reading here
), or the currency events that led to the wipeout of Long-Term Capital Management in 1998. The scientific literature works differently. As this paper shows, big events in organic chemistry don't come on as sudden, unexpected waves that sweep everything before them. Our swans are mute. They slip into the water so quietly that no one notices them for years, and they're often small enough that people mistake them for some other bird entirely. Thus the time lag.
How to shorten that? It'll be hard, because a lot of the dark-colored birds you see in the scientific literature aren't amazing black swans; they're crows and grackles. (And closer inspection shows that some of them are engaged in such unusual swan-like behavior because they're floating inertly on their sides). The sheer size of the literature now is another problem - interesting outliers are carried along in a flood tide of stuff that's not quite so interesting. (This paper mentions that very problem, along with a recommendation to still try to browse the literature - rather than only doing targeted searches - because otherwise you'll never see any oddities at all).
Then there's the way that we deal with such things even when we do encounter them. Nugent's recommendation is to think hard about whether you really know as much as you think you do when you try to rationalize away some odd report. (And rationalizing them away is the usual reponse). The conventional wisdom may not be as solid as it appears; you can probably put your foot through it in numerous places with a well-aimed kick. As the paper puts it: "Ultimately, the fact that something has never been done is the flimsiest of evidence that it cannot be done."
That's worth thinking about in terms of medicinal chemistry, as well as organic synthesis. Look, for example, at Rule-Of-Five type criteria. We've had a lot of discussions about these around here (those links are just some of the more recent ones), and I'll freely admit that I've been more in the camp that says "Time and money are fleeting, bias your work towards friendly chemical space". But it's for sure that there are compounds that break all kinds of rules and still work. Maybe more time and money should go into figuring out what it is about those drugs, and whether there are any general lessons we can learn about how to break the rules wisely. It's not that work in this area hasn't been done, but we still have a poor understanding of what's going on.
+ TrackBacks (0) | Category: Chemical News | Drug Industry History | The Scientific Literature | Who Discovers and Why
August 24, 2012
Lilly has reported results from its anti-amyloid antibody, solanezumab, and. . .well, it's mixed. And it's either quite good news, or quite bad. You make the call.
The therapy missed its endpoints (both "cognitive and functional", according to the company) in two clinical trials, so that's clearly bad news. Progression of Alzheimer's disease was not slowed. But I'll let the company's press release tell the tale from there:
The EXPEDITION1 study did not meet co-primary cognitive and functional endpoints in the overall mild-to-moderate patient population; however, pre-specified secondary subgroup analyses in patients with mild Alzheimer's disease showed a statistically significant reduction in cognitive decline. Based on those results, Lilly modified the statistical analysis plan (SAP) for EXPEDITION2 prior to database lock to specify a single primary endpoint of cognition in the mild patient population. This revised primary endpoint did not achieve statistical significance.
Now, this news - what you've just read above - actually is sending Lilly's stock up as I write this, which tells you how beaten-down Eli Lilly investors are, or how beaten-down investors in Alzheimer's therapies are. Or both. The headlines are all about how the drug missed in these trials, but that the company sees some hope. But man, is it ever a faint one.
What I'm taking away from the company's statement is that they had a cognition endpoint defined at the beginning of the trial (as well they should). We can assume that it was not a wildly optimistic one; no one is wildly optimistic in this field. And solanezumab missed it in the first Phase III data. But the patients with milder Alzheimer's, when they looked more closely, showed a trend towards efficacy, so they modified the endpoints (that is, lowered the bar and narrowed down to a select population) in the data for the second Phase III before it finished up. And even then, the antibody missed. So what we have are trends, possible trends, but nothing that really gets to the level of statistical significance.
But note, they're talking cognitive efficacy, and there's nothing said about those functional endpoints. If I'm interpreting this right, that means that there was a trend towards efficacy in tests like remembering words and lists of numbers, but not a trend when it came to actually performing better in real-life circumstances. Am I seeing this correctly? Lilly will be presenting more data in October, and we'll know more then. But I'm not getting an optimistic feeling from all this.
I assume that the company is now talking about going back and rounding up a population of the mildest Alzheimer's patients it can find and giving solanezumab another shot. Given Lilly's pipeline and situation, I suppose I'd do the same thing, but this is really a back-to-the-wall move. I think that you'd want to see something in a functional endpoint to really make a case for the drug, for one thing, and out in the real world, diagnosing Alzheimer's that early is not so easy, as far as I know. Good luck to them, but they are really going to need it.
+ TrackBacks (0) | Category: Alzheimer's Disease | Clinical Trials
Over at Chemistry Blog, there's a post by Quintus on the synthesis of a complex natural product, FR-182877. The route is interesting in that it features a key Diels-Alder reaction, and the post mentions that this isn't a reaction that gets used much in industry.
True enough - that one and the Claisen rearrangement are the first reactions I think of in the category of "taught in every organic chemistry course, haven't run one in years". In the case of the Claisen, the number of years is now getting up to. . .hmm, about 26, I think. The Diels-Alder has shown up a bit more often for me, and someone in my lab was running one last year, but it was the first time she'd ever done it (after many years of drug discovery experience).
Why is that? The post I linked to suggested a good reason that one isn't done too often on scale: it can be unpredictably exothermic, and some of the reactants can decide to polymerize instead, which you don't want, either. That can be very exothermic, too, and leaves you with a reactor full of useless plastic gunk which will have to be removed with tools ranging from a scoop to a saw. This is a good time to adduce the benefits of flow chemistry, which has been successfully applied in such cases, and is worth thinking about any time you have a batch reaction that might take off on you.
But to scale something up, you need to have an interest in that structure to start with. There's another reason that you don't see so many Diels-Alders in drug synthesis, and it has to do with the sorts of molecules we tend to make. The cycloaddition gives you a three-dimensional structure with stereocenters, and medicinal chemistry, notoriously, tends to favor flat aromatic rings, sometimes very much to its detriment. Many drug discovery departments have taken the pledge over the years to try to cut back on the flatness and introduce more sp3 carbons, but it doesn't always take. (For one thing, if your leads are coming out of your screening collection, odds are you'll be starting with something on the flat end of the scale, because that's what your past projects filled the files with).
I think that fragment-based drug discovery has a better chance of giving you 3-D leads, but only if you pay attention while you're working on it. Those hits can sometimes be prosecuted in the flat-and-aryl style, too, if you insist. And I think it's fair to say that a lot of fragment hits have an aryl (especially a heteroaryl) ring in them, which might reflect the ease of assembling a fragment-sized library of compounds full of such. Even the fragment folks have been talking over the years about the need to get more three-dimensionality into the collections, and vendors have been pitching this as a feature of their offerings.
The other rap on the classic Diels-Alder reaction is that it gives you substituted cyclohexanes, which aren't always the first place you look for drug leads. But the hetero-Diels-Alder reactions can give you a lot of interesting compounds that look more drug-like, and I think that they deserve more play than they get in this business. I'll go ahead and take a public pledge to run a series of them before the year is out!
+ TrackBacks (0) | Category: Chemical News | Life in the Drug Labs
August 23, 2012
So here's a comment to this morning's post on stock buybacks, referring both to it and my replies to Donald Light et al. last week. I've added links:
Did you not spend two entire posts last week telling readers how only pharma "knows" how to do drug research and that we should "trust" them and their business model. Now you seem to say that they are either incompetent or conmen looking for a quick buck. So what is it? Does pharma (as it exists today) have a good business model or are they conmen/charlatans out for money? Do they "know" what they are doing? Or are they faking competence?
False dichotomy. My posts on the Donald Light business were mostly to demonstrate that his ideas of how the drug industry works are wrong. I was not trying to prove that the industry itself is doing everything right.
That's because it most certainly isn't. But it is the only biopharma industry we have, and before someone comes along with a scheme to completely rework it, one should ask whether that's a good idea. In this very context, the following quote from Chesterton has been brought up, and it's very much worth keeping in mind:
In the matter of reforming things, as distinct from deforming them, there is one plain and simple principle; a principle which will probably be called a paradox. There exists in such a case a certain institution or law; let us say, for the sake of simplicity, a fence or gate erected across a road. The more modern type of reformer goes gaily up to it and says, "I don't see the use of this; let us clear it away." To which the more intelligent type of reformer will do well to answer: "If you don't see the use of it, I certainly won't let you clear it away. Go away and think. Then, when you can come back and tell me that you do see the use of it, I may allow you to destroy it."
This paradox rests on the most elementary common sense. The gate or fence did not grow there. It was not set up by somnambulists who built it in their sleep. It is highly improbable that it was put there by escaped lunatics who were for some reason loose in the street. Some person had some reason for thinking it would be a good thing for somebody. And until we know what the reason was, we really cannot judge whether the reason was reasonable. It is extremely probable that we have overlooked some whole aspect of the question, if something set up by human beings like ourselves seems to be entirely meaningless and mysterious. There are reformers who get over this difficulty by assuming that all their fathers were fools; but if that be so, we can only say that folly appears to be a hereditary disease. But the truth is that nobody has any business to destroy a social institution until he has really seen it as an historical institution. If he knows how it arose, and what purposes it was supposed to serve, he may really be able to say that they were bad purposes, that they have since become bad purposes, or that they are purposes which are no longer served. But if he simply stares at the thing as a senseless monstrosity that has somehow sprung up in his path, it is he and not the traditionalist who is suffering from an illusion.
The drug industry did not arise out of random processes; it looks the way it does now because of a long, long series of decisions. Because we live in a capitalist system, many of these decisions were made to answer the question "Which way would make more money?" That is not guaranteed to give you the best outcome. But neither is it, as some people seem to think, a guarantee of the worst one. Insofar as the need for new and effective drugs is coupled to the ability to make money by doing so, I think the engine works about as well as anything could. Where these interests decouple (tropical diseases, for one), we need some other means.
My problem with stock buybacks is that I think that executives are looking at that same question ("Which way would make more money?") and answering it incorrectly. But under current market conditions, there are many values of "wrong". In the long run, I think (as does Bruce Booth) that it would be more profitable, both for individual companies and for the industry as a whole, to invest more in research. In fact, I think that's the only thing that's going to get us out of the problems that we're in. We need to have more reliable, less expensive ways to discover and develop drugs, and if we're not going to find those by doing research on how to make them happen, then we must be waiting for aliens to land and tell us.
But that long run is uncertain, and may well be too long for many investors. Telling the shareholders that Eventually Things Will Be Better, We Think, Although We're Not Sure How Just Yet will not reassure them, especially in this market. Buying back shares, on the other hand, will.
+ TrackBacks (0) | Category: Business and Markets | Drug Development
Bruce Booth has an excellent look at a topic we were discussing around here earlier this year: stock buybacks in biopharma. I didn't have a lot of good things to say about the concept. I understand that corporations have obligations to their shareholders, and I certainly understand that a stock buyback is about the least controversial thing a big company can do with its money. Paying shareholders through dividends has tax consequences. But you can't sit on a big pile of cash forever, and what are you supposed to do if you think that market returns will beat the return on investment in your own business?
That brings up another, larger question: if you truly believe that last part, how long do you think that situation will obtain? And how long are you willing to put up with it? If a business really, truly, can't deliver returns that could be realized through a reasonable investment strategy, then why is it in business to start with? (I've seen discussions among economists about this very point when applied to many small businesses).
Booth wonders about the use of capital, too:
In recent years, plowing it back into internal R&D hasn’t been the preferred option given pipeline productivity questions. Returning capital to shareholders via dividends has certainly been high on the list. Another, albeit indirect, way of paying shareholders is through share repurchases (stock buybacks), and it has also been quite popular. The expectation (or hope) with these indirect stock buybacks is that the stock will move upwards because the shares oustanding goes down (or at least the buybacks offset the dilution from the exercise of options).
But buybacks have a more mixed assessment in practice (links at his site - DBL) and are typically only a smart if a company is (a) under-valued and (b) has no better uses of capital. This latter point is where they draw my ire, especially given their scale in our industry and the many strategic alternatives.
Totaling up the buybacks gives you some humongous figures. One thing that I'm not quite sure about with these numbers is whether all these buybacks are actually followed through. You'd think there would be legal consequences if the discrepancy grosw too large, but I don't know the law on this topic. But taking the figures as we have them, you get this:
To appreciate the magnitude of these buybacks, it’s worth comparing them to other important financial values in the biopharma ecosystem. It’s bigger than the NIH budget for both 2011-2012 by nearly 25%. It’s 4.5x bigger than all of the private venture-backed M&A that occurred in the past 18 months – and that involved over 70 biotech companies. It’s 12x bigger than the sum total of venture dollars invested in biotech in that period. And it's nearly 80x bigger than all the capital raised by fifteen biotech IPOs during that period. This is a huge amount of capital washing into stock repurchases.
The problem is, as Booth goes on to show, is that there's no particular correlation (that anyone can see) between these buybacks and the performance of the stocks themselves. (You could always say that they'd have performed even worse without the buybacks, an unanswerable and untestable point). He's got some other suggestions for the money, and he's not even asking for all of it. Or half of it. Or a tenth. Five per cent of the buyback pool would totally alter the funding universe for early-stage companies and precompetitive consortia. In other words, potentially alter the future of the whole industry. But we're not doing that. We're buying our own shares. Tens of billions of dollars of our own shares, because we can't seem to think of anything better to do.
+ TrackBacks (0) | Category: Business and Markets
August 22, 2012
Here's the Suzuki reaction taken down about as far as it can go: two boron groups on one carbon.
I didn't even know that you could make those things - no doubt someone will be inspired to try the three-boron version next. Diarylmethanes aren't the most preferred drug structures in the world (that carbon is just waiting to be oxidized), but I can't say that I've always avoided them on those grounds. I was on a project where we made a whole series of the things, actually - didn't work out so well for the intended target, but the compounds went on to hit in a completely different assay, so the company did probably get its money's worth.
+ TrackBacks (0) | Category: Chemical News
Hang around a bunch of medicinal chemists (no, really, it's more fun than you'd think) and you're bound to hear discussion of cLogP. For the chemists in the crowd, I should warn you that I'm about to say nasty things about it.
For the nonchemists in the crowd, logP is a measure of how greasy (or how polar) a compound is. It's based on a partition experiment: shake up a measured amount of a compound with defined volumes of water and n-octanol, a rather greasy solvent which I've never seen referred to in any other experimental technique. Then measure how much of the compound ends up in each layer, and take the log of the octanol/water ratio. So if a thousand times as much compound goes into the octanol as goes into the water (which for drug substances is quite common, in fact, pretty good), then the logP is 3. The reason we care about this is that really greasy compounds (and one can go up to 4, 5, 6, and possibly beyond), have problems. They tend to dissolve poorly in the gut, have problems crossing membranes in living systems, get metabolized extensively in the liver, and stick to a lot of proteins that you'd rather they didn't stick to. Fewer high-logP compounds are capable of making it as drugs.
So far, so good. But there are complications. For one thing, that description above ignores the pH of the water solution, and for charged compounds that's a big factor. logD is the term for the distribution of all species (ionized or not), and logD at pH 7.4 (physiological) is a valuable measurement if you've got the possibility of a charged species (and plenty of drug molecules do, thanks to basic amines, carboxylic acids, etc.) But there are bigger problems.
You'll notice that the experiment outlined in the second paragraph could fairly be described as tedious. In fact, I have never seen it performed. Not once, and I'll bet that the majority of medicinal chemists never have, either. And it's not like it's just being done out of my sight; there's no roomful of automated octanol/water extraction machines clanking away in the basement. I should note that there are other higher-throughput experimental techniques (such as HPLC retention times) that also correlate with logP and have been used to generate real numbers, but even those don't account for the great majority of the numbers that we talk about all the time. So how do we manage to do that?
It has to do with a sleight of hand I've performed while writing the above sections, which some of you have probably already noticed. Most of the time, when we talk about logP values in early drug discovery, we're talking about cLogp. That "c" stands for calculated. There are several programs that estimate logP based on known values for different rings and functional groups, and with different algorithms for combining and interpolating them. In my experience, almost all logP numbers that get thrown around are from these tools; no octanol is involved.
And sometimes that worries me a bit. Not all of these programs will tell you how solid those estimates are. And even if they will, not all chemists will bother to check. If your structure is quite close to something that's been measured, then fine, the estimate is bound to be pretty good. But what if you feed in a heterocycle that's not in the lookup table? The program will spit out a number, that's what. But it may not be a very good number, even if it goes out to two decimal places. I can't even remember when I might have last seen a cLogP value with a range on it, or any other suggestion that it might be a bit fuzzy.
There are more subtle problems, too - I've seen some oddities with substitutions on saturated heterocyclic rings (morpholine, etc.) that didn't quite seem to make sense. Many chemists get these numbers, look at them quizzically, and say "Hmm, I didn't know that those things sorted out like that. Live and learn!" In other words, they take the calculated values as reality. I've even had people defend these numbers by explaining to me patiently that these are, after all, calculated logP values, and the calculated log P values rank-order like so, and what exactly is my problem? And while it's hard to argue with that, we are not putting our compounds into the simulated stomachs of rationalized rodents. Real-world decisions can be made based on numbers that do not come from the real world.
+ TrackBacks (0) | Category: Drug Assays | In Silico | Life in the Drug Labs
August 21, 2012
This paper from GlaxoSmithKline uses a technology that I find very interesting, but it's one that I still have many questions about. It's applied in this case to ADAMTS-5, a metalloprotease enzyme, but I'm not going to talk about the target at all, but rather, the techniques used to screen it. The paper's acronym for it is ELT, Encoded Library Technology, but that "E" could just as well stand for "Enormous".
That's because they screened a four billion member library against the enzyme. That is many times the number of discrete chemical species that have been described in the entire scientific literature, in case you're wondering. This is done, as some of you may have already guessed, by DNA encoding. There's really no other way; no one has a multibillion-member library formatted in screening plates and ready to go.
So what's DNA encoding? What you do, roughly, is produce a combinatorial diversity set of compounds while they're attached to a length of DNA. Each synthetic step along the way is marked by adding another DNA sequence to the tag, so (in theory) every compound in the collection ends up with a unique oligonucleotide "bar code" attached to it. You screen this collection, narrow down on which compound (or compounds) are hits, and then use PCR and sequencing to figure out what their structures must have been.
As you can see, the only way this can work is through the magic of molecular biology. There are so many enzymatic methods for manipulating DNA sequences, and they work so well compared with standard organic chemistry, that ridiculously small amounts of DNA can be detected, amplified, sequenced, and worked with. And that's what lets you make a billion member library; none of the components can be present in very much quantity (!)
This particular library comes off of a 1,3,5-triazine, which is not exactly the most cutting-edge chemical scaffold out there (I well recall people making collections of such things back in about 1992). But here's where one of the big questions comes up: what if you have four billion of the things? What sort of low hit rate can you not overcome by that kind of brute force? My thought whenever I see these gigantic encoded libraries is that the whole field might as well be called "Return of Combichem: This Time It Works", and that's what I'd like to know: does it?
There are other questions. I've always wondered about the behavior of these tagged molecules in screening assays, since I picture the organic molecule itself as about the size of a window air conditioner poking out from the side of a two-story house of DNA. It seems strange to me that these beasts can interact with protein targets in ways that can be reliably reproduced once the huge wad of DNA is no longer present, but I've been assured by several people that this is indeed the case.
In this example, two particular lineages of compounds stood out as hits, which makes you much happier than a collection of random singletons. When the team prepared a selection of these as off-DNA "real organic compounds", many of them were indeed nanomolar hits, although a few dropped out. Interestingly, none of the compounds had the sorts of zinc-binding groups that you'd expect against the metalloprotease target. The rest of the paper is a more traditional SAR exploration of these, leading to what one has to infer are more tool/target validation compounds rather than drug candidates per se.
I know that GSK has been doing this sort of thing for a while, and from the looks of it, this work itself was done a while ago. For one thing, it's in J. Med. Chem., which is not where anything hot off the lab bench appears. For another, several of the authors of the paper appear with "Present Address" footnotes, so there has been time for a number of people on this project to have moved on completely. And that brings up the last set of questions, for now: has this been a worthwhile effort for GSK? Are they still doing it? Are we just seeing the tip of a large and interesting iceberg, or are we seeing the best that they've been able to do? That's the drug industry for you; you never know how many cards have been turned over, or why.
+ TrackBacks (0) | Category: Chemical Biology | Chemical News | Drug Assays | Drug Industry History
There's no telling if this is true - it's part of a lawsuit. But a former Genentech employee is claiming that the company rushed trials of its PI3K inhibitor. And why? Worries about their partner:
The suit alleges that the Pi3 Kinase team was guilty of "illegal and unethical conduct" by skirting established scientific and ethical standards required of drug researchers. Juliet Kniley claims she complained in 2008 and then was sidelined in 2009 with a demotion after being instructed to push ahead on the study. And she says she was told twice that Roche would "take this molecule away from us" if they saw her proposed timelines.
Genentech denies the allegations. But you have to wonder if there's still a window here into the relationship between the two companies. . .
+ TrackBacks (0) | Category: Drug Development
August 20, 2012
The controversy I wrote about last week, about whether (some) enzymes work by using extremely fast movements (rather than by putting things into their place and letting them do their thing) may remind some folks of the supposed medieval arguments about angels dancing on the heads of pins. But it also reminds me a bit of some other arguments in organic chemistry over the years. The horrible prototype is, of course, the norbornyl cation.
There was a time when people would simply leave the room when that topic came up, because they knew that they were in for another round of fruitless wrangling. Was its structure that of two rapidly interconverting standard carbocations, or a single bridged "non-classical" one that broke the previously accepted rules? George Olah and H. C. Brown, Nobel laureates both, were on opposite sides of that one, but every physical organic chemist from about 1950 to about 1980 probably had to take a stand one way or the other. (It is commonly accepted that Olah's side won), but the arguments got pretty esoteric by the end. Update: the battle was first joined by Saul Winstein, who did not live to see his proposal vindicated by Olah's spectroscopic studies).
Another one, which came along a few years later, was the "synchronous / asynchronous" mechanism of the Diels-Alder reaction. Do the new bonds in that one form at the same time, or does one form, and then the other? That one involved the physical organic people again, as well as plenty of computational chemists. I stopped following the debate after a while, but I believe that the final reckoning was that most standard Diels-Alder reactions were synchronous, within the limits of detection, but that messing with the electron density of the two reactants could easily push the reaction into asynchronous (or flat-out stepwise) territory.
So why does this level of detail matter? The problem is, chemistry is all about things like bond formation and bond breaking, and about interactions between individual molecules (and parts of molecules) that change the energies of the systems involved. And those things are nothing but picky details, all the way down. Thermodynamics, which runs chemical reactions and runs the rest of the universe, is the most rigorous branch of accounting there is. Totaling up those energies to see which side of the ledger wins out can easily involve the fate of single water molecules, or even to single protons, and you don't get much pickier than that.
This sort of thing is one argument used against the feasibility of molecular nanotechnology. How are we to harness such fine distinctions, at such levels? But it's worth remembering that we ourselves, and every other living creature, are nanotech machines at heart. Our enzymes are constantly breaking bonds, twisting single molecules, altering reaction rates, and generating specific, defined molecular products. If they weren't, we'd fall right over. We eventually fall over anyway, because none of these machines work perfectly. But they work pretty well, and they make our own chemical efforts look like stone axes and deer-bone hammers.
So we may find getting down to this level of things to be a lot of work, and hard to understand, and frustrating to deal with. But that's where we're going to have to be if we're ever going to do real chemistry, the kind that's that's indistinguishable from magic.