Skip to main content
Menu

Business and Markets

Peter Thiel’s Uncomplimentary Views of Big Pharma

See what you think of Peter Thiel’s characterization of the drug industry in this piece for Technology Review. Thiel’s a very intelligent guy, and his larger points about technology stalling out make uncomfortable reading, in the best sense. (The famous quote is “We wanted flying cars; instead we got 140 characters”). But take a look at this (emphasis added):

You have to think of companies like Microsoft or Oracle or Hewlett-Packard as fundamentally bets against technology. They keep throwing off profits as long as nothing changes. Microsoft was a technology company in the ’80s and ’90s; in this decade you invest because you’re betting on the world not changing. Pharma companies are bets against innovation because they’re mostly just figuring out ways to extend the lifetime of patents and block small companies. All these companies that start as technological companies become antitechnological in character. Whether the world changes or not might vary from company to company, but if it turns out that these antitechnology companies are going to be good investments, that’s quite bad for our society.

I’d be interested in hearing him revise and extend those remarks, as they say in Washington. My initial reaction was to sit down and write an angry refutation, but I’m having second thoughts. The point about larger companies becoming more cautious is certainly true, and I’ve complained here about drug companies turning to M&A and share buybacks instead of putting that money back into research. I’d say, though, that the big drug companies aren’t so much anti-technology as they are indifferent to it (or as indifferent as they can afford to be).
Even that still sounds harsh – what I mean is that they’d much rather maximize what they have, as opposed to coming up with something else. Line extensions and patent strategies are the most obvious forms of this. Buying someone else’s innovations comes next, because it still avoids the pain and uncertainty of coming up with your own. There’s no big drug company that does only these things, but they all do them to some degree. Share buybacks are probably the most galling form of this, because that’s money that could, in theory, be applied directly to R&D, but is instead being used to prop up the share price.
But Thiel mentions elsewhere in his interview that we could, for example, be finding cures for Alzheimer’s, and we’re not. Eli Lilly, though, is coming close to betting the company on the disease, taking one huge swing after another at it. Thiel’s larger point stands, about how more of the money that’s going into making newer, splashier ways to exchange cat pictures and one-liners over the mobile phone networks could perhaps be applied better (to Alzheimer’s and other things). But it’s not that the industry hasn’t been beating away on these itself.
I worry that the Andy Grove fallacy might be making an appearance again, given Thiel’s background (PayPal, Facebook, LinkedIn). That link has a lot more on that idea, but briefly, it’s the tendency for some people from the computing/IT end of the tech world to ask what the problem is with biomedical research, because it doesn’t improve like computing hardware does. It’s a good day to reference the “No True Scotsman” fallacy, too: sometimes people seem to identify “technology” with computing, and if something doesn’t double in speed and halve in cost every time you turn around, well, that’s not “real” technology. At the very least, it’s not living up to its potential, and there must be something wrong with it.
I also worry that Thiel adduces the Manhattan project, the interstate highway system, and the Apollo program as examples of the sort of thing he’d like to see more of. Not that I have anything against any of those – it’s just that they’re all engineering projects, rather than discovery ones. The interstate system, especially: we know how to build roads, so build bigger ones. The big leap there was the idea that we needed large, standardized ones across the whole country, with limited entrances and exits. (And that was born out of Eisenhower’s experiences driving across the country as the road network formed, and seeing Germany’s autobahns during the war).
But you can say similar things about Apollo: we know that rockets can exist, so build bigger ones that can take people to the moon and back. There were a huge number of challenges along the way, in concept, design, and execution, but the problem was fundamentally different than, say, curing Alzheimer’s. We don’t even know that Alzheimer’s can be cured – we’re just assuming that it can. I really tend to think it can be cured, myself, but since we don’t even know what causes it, that’s a bit of a leap of faith. We’re still making fundamental “who knew?” type discoveries in biochemistry and molecular biology, of the sort that would totally derail most big engineering projects. The Manhattan project is the closest analog of the three mentioned, I’d say, because atomic physics was such a new field (and Oppenheimer had to make some massive changes in direction along the way because of that). But I’ve long felt that the Manhattan project is a poor model, since it’s difficult to reproduce its “Throw unlimited amounts of money and talent at the problem” mode, not to mention the fight-for-the-survival-of-your-civilization aspect.
But all that said, I do have to congratulate Peter Thiel on putting his money down on his ideas, though his investment fund. One of things I’m happiest about in today’s economy, actually, is the way that some of the internet billionaires are spending their money. Overall, I’d say that many of them agree with Thiel that we haven’t discovered a lot of things that we could have, and they’re trying to jump-start that. Good luck to them, and to us.

59 comments on “Peter Thiel’s Uncomplimentary Views of Big Pharma”

  1. Hap says:

    Considering Retrophin, though, maybe it’s not (only) bigger pharma companies that are trying to make money by preventing other people from selling useful things.

  2. John Galt III says:

    I recently commented that if I were a billionaire, I’d be spending money on space exploration and asteroid deflectors. I like that they are doing at least some of what I’d do.

  3. Electrochemist says:

    Thiel suffers from an advanced case the “all or nothing” disease common to many who dress in black and spend their days sipping soy lattes and watching TedX talks on YouTube.
    Pharma companies are not anti-technology any more than car companies, oil companies, the government, NIH, the food industry, etc., are anti-technology. We are simply not early adopters, and cannot jump on the latest fad, and reorganize every 6 months. It isn’t the business of large organizations to be “on the cutting edge.”
    There are good and bad examples to be had in every industry. The best large organizations maintain a robust network of outside partners, and bring in new technologies once the risk/benefit ratio can be at least somewhat understood. Their main focus, of course, is to bring (other people’s) innovation to the masses (including spending $500M per molecule over a decade or more to commercialize and register new medicines).

  4. I agree with @Electrochemist. Thiel is comparing very different things. It’s one thing to build a web site or service or an app with $1 million and a room full of coders. It’s quite another to delay Alzheimer’s progression. Thiel may be a smart guy, but he doesn’t know what he’s talking about here.

  5. Anonymous says:

    I don’t think Theil is comparing the relative ease of engineering compared to Alzheimer’s, just that investors would rather have a whats app than do anything useful for humanity.

  6. Anon says:

    Thiel has something like Nobel disease:
    “The Nobel disease is a term used to describe a phenomenon in which Nobel Prize-winning scientists endorse or perform “research” in pseudoscientific areas in their later years. In reality, this “disease” most likely demonstrates that even the most brilliant people are not immune to crank ideas and belief in such ideas will persist to some degree even among Nobelists. It also makes for a convenient argument from authority for lesser cranks, because if a Nobel Prize-winning scientist says it, it must be true”
    Just because he made money investing in social media, doesn’t make him an expert in investing in pharma/biotech and also doesn’t mean he is an expert at how those companies should be run.

  7. Rhenium says:

    After the debacle with Thiola from last week, it’s not just big companies either…

  8. Big Farmer says:

    Reality is that just as with tech, pharma takes the path of least resistance and ends up with reformulations and drugs in pathways they understand (akin to 140 characters instead of flying cars). Pharma needs to dig in and figure out how to take the road less traveled.

  9. John Wayne says:

    Everybody wants a flying car, the cure for all diseases, and a way to stop aging. As a society, we are beginning to understand how to make true game-changing events happen (money, time, people), but most people have no idea what the heck is going on.
    Maybe we should tell people that the folks who invent all the new drugs at the NIH are slacking off. Those of us in pharma have nothing to steal, leading to a lack of innovation. A commitment to funding more basic research would help a lot. This isn’t the ‘truth,’ but it is way easier to explain than telling people just how ignorant we are.

  10. Hap says:

    It’s going to be easier to make apps than drugs, and entertainment has few limits on what you can charge for your product (just because I want an IPad doesn’t mean Apple has to give one cheaply – I can live without one easily), while drugs and other important things do (in law or in practice). Unless that were going to change (which I don’t think society is willing to contemplate), what we invest in to make money and what we think society should invest in are two very different things.

  11. Bob Warfield says:

    If the big companies are in fact using patents, blocking smaller companies, and doing share buybacks, then they are most assuredly anti-technology. To prevent others from making progress or to actively choose not to invest is to be anti. It is to be not a company of scientists but a company of lawyers, accountants, and MBA’s.
    Thiel shouldn’t be throwing stones as all of his successes had no technology of note to speak of–just off the shelf obvious stuff. Many have argued that Silicon Valley has been way too focused on ventures like Thiel’s to the detriment of creating much new technology.
    But there have been exceptions and the biggest one goes to the heart of your so-called “Andy Grove Fallacy.” There is no Andy Grove Fallacy–it’s hubris and a lack of understanding of what people like Grove overcame. You draw the 100,000 foot high distinction between building microprocessors and understanding biological systems as though the two have nothing in common, but in fact they do.
    The chip industry spent decades understanding the physics underlying phenomena at those scales well enough to manufacture reliable devices. Take a photo of a chip fab clean room and explain to me why it is any less profoundly scientific than a photo of a chem lab or a NASA satellite assembly room.
    What’s different is the big players invested hugely in that technology instead of M&A and the other things Big Pharma is doing today. Perhaps Big Pharma has swung for their fence, missed, and given up. No harm no foul, but to therefore say essentially, “Aw the chip and computer guys had it easy and our job is so much harder they can’t even talk,” well, if you want to shut them up, you’re going to have to put on a better show than Andy Grove put downs.
    Figure it out, the world is depending on. Don’t cop out and throw stones at the other guys, there’s no progress in that.

  12. Hap says:

    No, the “Andy Grove Fallacy” is real – you missed the point. Engineered systems were designed by people – they may have had to figure things out to make them but the underlying design was human in intention and design, and can be understood by humans accordingly. Biological systems don’t have such design parameters – they may not have any “design” parameters, but stochastic processes acting under selection. You can’t rely on purpose or intent to figure them out, and you may not even be able to know when you know all of the system parameters. They behave differently, and knowledge applied to one does not necessarily apply in the other. (Biological systems still rely on physics and chemistry and thermo – they can’t escape those, and knowing them is useful, just that there are lots of ways to use those to run a biological system). The assumption that underlies Grove’s complaints is fallacious.
    This, however, doesn’t mean Pharma (in general) or its investors haven’t given up – it just means that the magnitude of the problems they face is substantially different than those faced in engineering fields.

  13. johnnyboy says:

    @11 – you don’t get it at all, as is made clear by your statement “Take a photo of a chip fab clean room and explain to me why it is any less profoundly scientific than a photo of a chem lab or a NASA satellite assembly room.”, by which you demonstrate that you do not understand the difference between engineering and science. Science is not about fancy shiny techie-looking things; science is about understanding, you can’t take a pretty picture of it, because it mainly goes on in the brains of researchers, not in satellite assembly rooms.
    The chip industry may have done a lot of research in “understanding the physics underlying the phenomena”, and that’s great, because it meant that said physics became understood well enough for them to engineer their widgets better. The difference with life sciences is that the “phenomena” for most diseases are so complex that we do not understand them fully, if at all – certainly not to the point where we can engineer perfect drugs.
    That doesn’t mean we aren’t trying or putting enough resources in. You can see here that pharma and biotech has the highest investment in R&D of all industries:
    http://europa.eu/rapid/press-release_MEMO-12-948_en.htm

  14. daveh says:

    I’m usually in agreement with you however,: “Pharma companies are bets against innovation because they’re mostly just figuring out ways to extend the lifetime of patents and block small companies” is a reality based statement or we wouldn’t be considering this legislation.
    http://blogs.wsj.com/pharmalot/2014/09/19/legislation-would-prevent-drug-makers-from-thwarting-generic-rivals/
    I guess I would’t have used “mostly”, but the race is neck and neck with NCE development.

  15. Bell4 says:

    Re @11, Hap has again nailed it.
    Mr/Dr Warfield – simple question: please tell us whether the beta-secretase inhibitors will work in Alzheimer’s. Be sure to specify the percent improvement (if there is any) that will be observed in the trials that will come out in a few years.
    Answering this question is orders of magnitude easier than dealing with many other puzzles of human biology. Sadly, those of us laboring in the field, ignorant wretches that we are, just can’t come up with a robust answer without doing horribly expensive and time-consuming human trials. Meanwhile, investors are tapping their foot impatiently and looking for some other place to put their capital.
    What’s that you say? We haven’t done the right science to answer to this question because we’re all such dolts? Silly me, I’m sure you’re right. Just go ahead and write down the science we should be doing to address this properly. Make sure to clarify how existing technology will get us there, or go ahead and invent some new stuff – either is good.
    We’re waiting.

  16. Brett says:

    That’s a good comparison with the Apollo Program. We knew how the physics of conventional rocketry would work – hell, we had the Rocket Equation decades before we even had rockets capable of getting into space! We just didn’t have the engineering to get there, or know if the engineering was there to discover.

  17. ron says:

    I’d say you are not too far off of what Theil is saying. ‘Anti-technology’ doesn’t mean torches and pitchfork opposition. The same ambivalence you describe for big Pharma is the same attitude that OldTech has. They don’t hate innovation, its just not central to their goals or their strategy. No Moon shot (or cancer cure) is going to happen with an attitude like that.
    As far as engineering vs science goes that distinction is overdrawn. Far too few scientists care how to make their findings work and could learn some lessons from engineering. Tech is neither science nor engineering it seems to me.

  18. Bob Warfield says:

    Folks, the fact that chips are human designed and organisms are not is a Red Herring. You’ve mismatched the components of the analogy. The chip is the drug–both are human designed in some sense. The complexity is the underlying physics in the chip’s case and the biology/biochemistry in the organism’s case.
    Match them up that way and we’re back to no Grove Fallacy. Think about it that way and you’ve got a shot at changing how the Pharma game is played. Chip design and electronics at those scales was a total black art when it started. People had no idea why various phenomena were occurring or why things that seemed obvious didn’t work. But, to a greater degree than pharma has, they insisted on understanding the science analytically and not empirically. So there were many breakthroughs in solid state physics and having those breakthroughs delivered tremendous advantage.
    If nothing else, data was far more systematically obtained. Every time I come in contact with the medical world I see two things:
    – There’s a lot they don’t know about the complex non-linear zillion variable systems we call humans.
    – Their ability to gather the vast amounts of data needed to reverse engineer that system are lousy compared to what we see in most other fields. Records on paper, time spent re-entering the same basic information over and over again, very little ability to aggregate it and understand the inter-relationships. It’s a mess and the field is only in the very early stages of addressing the issue.
    This blog likes to talk about the problems of Pharma–Big Pharma’s faults, the war between the public’s perception of drug pricing and the realities of the industry, yada, yada. All that really boils down to the issue that is ideally suited for a Manhattan Project style approach–drug discovery is too haphazard to be called a business as is drug testing.
    And that, Bell4, is why nobody can tell you what your drug’s impact on Alzheimer’s will be. To put it in more straightforwardly, we’re too busy testing all these ideas in these specific situations without dropping back to figure out how the actual parts all fit together. That’s what the chip guys and many of these other disciplines did.
    You can’t reverse engineer complex systems by random highly specific experiments that are not coordinated or aimed at the overall goal, though you will sometimes discover something useful at great expense by doing so. And that’s what Pharma does.
    Does it have to be the only thing? Heck no. When someone gets motivated to do it more like the other tech areas operate, you get things like the Human Genome Project that happen far faster and more cheaply than anyone would project. But if everyone calling the shots is an MBA or someone whose best years of research thinking were 20 years ago, they’re just not going to make those investments. They’re too short-term oriented.

  19. anon says:

    Trying to develope drugs is sort of like trying to engineer a microchip in a world where new elementary particles, interactions and forces are being discovered on a weekly basis…

  20. Bob Warfield says:

    @johnnyboy, you make my point eloquently. Yes, there is a vast difference between science and engineering. What the chip guys did that pharma has not is they realized they were never going to get far until they could turn the science into engineering, and they paid the high price to do that gladly.
    That’s what pharma has largely failed to do. They delight in continuing to do science. Each drug is a new discovery, but for the most part that sort of discovery contributes little to making the science into engineering. The discoveries and research need to be focused on the latter.
    If you spend all your time wondering about specific apples falling from specific trees, and you travel the world checking every possible tree, apple, local weather condition, and so on, you will never arrive at any general theories of gravity.

  21. Hap says:

    A chip is part of a computer – the system in which it fits in is fully designed. You know what the inputs are, you know what the outputs are, and you know what part of the inputs are relevant (they all are). It is a closed system, and its relationships with other parts of the system are also defined by people. You can improve parts of it individually with lots of work without necessarily changing the whole thing. The unknowns are known, and can be found with lots of work but with some predictability.
    Contrast to drugs. People test drugs in parts of a system and in the full system (people) and get drastically different results. We don’t know how the parts all fit together, and which ones interact with others, so that local tasks may not fit with organism-wide ones. The problems involved with making a drug aren’t separable – where the lack of engineering takes importance. Unlike engineered things, the parts don’t seem to fit in separable ways because of the lack of intentional design. The unknowns aren’t known. It is worth spending lots to learn about, and not worth giving up on, but the dimensions of knowledge needed to delineate it are significantly more numerous, and significantly less predictable.

  22. RKN says:

    tendency for some people from the computing/IT end of the tech world to ask what the problem is with biomedical research, because it doesn’t improve like computing hardware does
    If these people had to make their technology work in a mouse before they were permitted to make it available in the Play Store, they’d begin to understand what the problem is.

  23. Moses says:

    The three examples of big engineering projects which Thiel uses, and to which you refer, had one thing in common: they were all state-funded, not by private industry.
    Contractors did the work, to be sure, but government funded it. Perhaps we should increase the role of the state in basic medical research again. After all, one of the Human Genome Project routes was state funded and it worked as well as Venter’s.

  24. Anon2 says:

    @ Bob Warfield
    I see you’ve singled out the hardware industry when the original post was speaking about software development, but I’ll continue on your path.
    You seem to be missing the different in engineering and science.
    -In engineering you start with macro scale assumptions and try to reduce them to practice on smaller and smaller scales. You can not do this with drug discovery, period.
    -In engineering you start with a given set of properties that you would like to have in a product. This product could be the pushing as many calculations as possible while staying within a certain temperature and using a certain amount of energy. You can’t do that in drug development as the biology isn’t well known. We still have reason to question many “accepted” biomarkers. Hardware started with motors and worked all the way down to processors. Each time in order to make an incremental advance you would try to say…create a material that can take higher temps. In doing so you make new allows or study physics and figure out better ways to do this. In doing so you get something incrementally better than the previous generation. If this were the case for the drug industry we would have all started with aspirin and have been slowing making modifications over the last 50 years and arrived at some incredible antiplatelet drug. This can’t be done and isn’t done for a number of reasons. The most obvious is that in hardware you can intuitively and/or even computationally make adjustments to a design and have a good idea of the outcome. In biology you can not add a Fluorine to aspirin and know what is going to happen.
    -In engineering each discovery becomes a new rule that is used to develop the next generation of system. In Biology you are looking at the nano, microscopic, and macroscopic views at the same time and since all is not known at all of these levels we can not simply create rules to engineer with. If you suddenly learn of a mutation in an Alzheimer’s gene, that doesn’t mean that you now have a new rule to add to the playbook when developing a new drug. I just means that you may know why something might be happening under certain conditions in a certain type of cellular system.
    Also “- Their ability to gather the vast amounts of data needed to reverse engineer that system are lousy compared to what we see in most other fields. Records on paper, time spent re-entering the same basic information over and over again, very little ability to aggregate it and understand the inter-relationships. It’s a mess and the field is only in the very early stages of addressing the issue.”
    It is harder to gather data in “their” field than it is in hardware. Also, I don’t know what you mean about paper record and re-entering information over and over. Are you about to try to sell us on another Google-esque “our computers can solve cancer better than you can” sales pitch? I’m only kidding here, of course.

  25. David Borhani says:

    @18, Bob Warfield: “Chip design and electronics at those scales was a total black art when it started. People had no idea why various phenomena were occurring or why things that seemed obvious didn’t work. But, to a greater degree than pharma has, they insisted on understanding the science analytically and not empirically.”
    I disagree. The “analytical” understanding of how semiconductors work dates back to “science” (quantum mechanics, associated solid-state physics, and the chemistry of group III, IV, and V elements) carried out in the 1920s-1950s.
    Conversely, understanding why certain ways of baking a bit of Al or P into a microscopic slab of Si (and indeed, even how to purify that Si to 99.999999…% purity) was *engineering*. Let’s vary the obvious, and probably a few not-so-obvious (at least at first) parameters: temperature, annealing schedule, partial pressure, exposure time, Al/P source compound, yada, yada, yada. All things understood to likely (or at least potentially) be important before even doing the systematic try-all-variations approach.
    I think @19, anon has it right. How well would chip design have improved from 1960 to present if the underlying characteristics of chips depended, in critical and non-linearly complex ways, upon neutrinos oscillating between 3 (no, let’s make that 10,000) different flavors? And posit that neutrinos weren’t discovered in the 1930s, but that they will only become known (but not yet fully understood) until 10 years from now.
    The Andy Groves fallacy does unfortunately exist — in essence, it is the hubris of the engineer who forgets the decades or centuries of scientists who preceded him (or her).
    That’s not to say that some early engineers (architects) sometimes just got lucky: think Brunelleschi and his dome. My guess is that he didn’t understand one wit of the physics that would allow his dome to withstand both gravity and the test of time, but he must have had an excellent intuitive feel for it might work (not unlike some designers of drugs?), and he apparently also played with some much smaller models first.
    In spite of all this, I think you DO make a good point: Understanding general behavior and rules, where possible, will of course be of immense help. Many people, in Pharma and outside of it, are trying to do just that.

  26. Harrison says:

    @Bob Warfield. I think Hap did a good job of explaining where the parallels fall apart, but I wanted to add to that.
    If a chip is a drug, then imagine having to design your chip in isolation and then test the chip on hundreds of different motherboards by different manufacturers to figure out if your chip was compatible. While there are undoubtedly mysteries in chip manufacturing, they are variables in a designed system. Organisms have such random and haphazard features at literally every step (think of the human eye and visual processing compared to a camera/computer) that no amount of reductionism will ever equal the understanding of a microchip.

  27. Anonymous says:

    One aspect of cells, organs, individuals is that they are evolved systems, not “clean sheet of paper” designs such as microprocessors. Evolution takes pathways that are not obvious, and the changing environment also leaves an impact. We are still very much learning about how the cell’s systems work together, to say nothing about integrated systems such as organs and organisms.
    Having spent lots of time in 3 big pharmas, I must say that the managerial attitudes (influenced by Wall Street and the ever popular “Shareholder value”) have changed from a more innovative approach to a more evergreening philosophy. Too, we made mistakes by over investing in combichem and HTS of random chemical libraries; neither of which have really panned out in a big way.

  28. Bell4 says:

    @11, @20 – the other commenters have nailed this one, but I have to ask, do you actually know anything about drug discovery/biology (i.e., have a degree in some related field, or have done sustained research in same), or are you just today’s illustration of the Dunning-Kruger Effect?

  29. entrepreneur says:

    Please let me know which VC would fund a project like flying cars which is laden with:
    1 high regulatory burden, perhaps EPA, NHTSA, NTSB, FAA?
    2 capital intensive R&D
    3 long long development time
    4 risk of explosion/death
    I ask because I want to ask them to fund my drug discovery biotech!

  30. gippgig says:

    Biomedical research doesn’t improve like computing hardware does? If computers had improved at the same rate as DNA sequencing computers would have thousands of GB of RAM and run at millions of GHz.
    Alzheimer’s can almost certainly be halted or prevented but it is questionable whether it can be cured – if a destroyed mind can be restored are they still the same person?

  31. Issues of engineering versus science aside, the underlying systems which must be understood are orders of magnitude different in complexity. It’s all very well to say medicine should be more organized in basic science, like the semiconductor guys were. There might be 10 orders of magnitude more complexity in the biological system.
    The engineering problem of managing the data necessary to truly understand human biology at that level might itself be intractable. And that’s just the meta problem to get started.
    If it is, we’re stuck with heuristics, which is basically what we do now. I work in computing, on intractable problems. Not the fiddly little stuff google and NOAA work on, really intractable stuff.
    I’m pretty sure biology is much much harder than the problems my group hacks miserably away at.

  32. mike says:

    Thiel’s experience investing is in a field where “disruptive innovation” takes the form of creating something that people want that they never even knew they needed. Healthcare is different. Diabetics know they need medicine to lower their glucose levels. Cancer patients know they need treatment to rid the cancer from their bodies. We know they need these things too. The work is hard, long and not for the faint of heart. And it can never be done by a room full of 23 year olds living in a house together for a year fueled by pizza and mountain dew.

  33. MoMo says:

    Ye Gods! Another buzzword-disruptive! Billionaires without chemical or drug discovery biology experience let alone drug design are all too common-thinking its SO simple-like coding apps or creating interactive websites for the masses.
    History is repeating itself. Bill Gates did this in 1999 with the Medicines for Malaria Venture. Hired the best scientists, funded progressive CHEMICAL research- and what have they discovered in 16 years? That a drug from China-the artemisinins and derivatives and used for decades if not longer-are still the most potent drug useful today. NCEs are still in preclinical but None have been the magic bullet they hoped THE DISRUPTIVE model would find.
    So go ahead DISRUPTERS! DISRUPT AWAY!
    You ‘ll be wishing you had your hands in a big bowl of Intel chips again or writing programs that stream cats playing piano faster!

  34. aa3 says:

    I really agree with Hap.. In tech industries, aka the open areas of the economy that developed in recent times, people are allowed to charge whatever they want for their products.
    Its why I full on support Gilead’s pricing, and even think it should be higher. If people think the drug is too expensive for the benefit, well they can choose not to buy it. There is this popular belief that medicine is too important to be left to capitalism and property rights. I view it, that in medicine it is even more important that our societies adhere to property rights, as the price of not having treatments is so terrible for the people who need them.
    An analogy is the Soviet Union viewed farming as too important to allow farmers to profit from it, that people’s need for food should not be left in the hands of greedy family farmers. So they nationalized farming and took away food surpluses from farmers.. we know the results of that.
    If investment in innovation is slowing down, it is obvious to any well read person that the rewards are not high enough to justify the investments.

  35. Nick K says:

    Just to add a little to the debate about the Andy Grove Fallacy….silicon chips are strictly hierarchical, compartmentalised, and modular. Complex they may be, but errors and faults in them can be repaired because they can be isolated and thus understood. Biological systems are basically kludges lacking hierarchy and modularity, and are insanely complex and interconnected. Intervention in one place frequently causes unexpected effects elsewhere (Sildenafil anyone? PDE 5 inhibition does different things in different places). It’s actually surprising that we have any drugs at all.

  36. Anonymous says:

    Sorry for plagiarizing myself but here is a repeat of an earlier comment from a similar thread:
    The main problem,IMHO, with the people trying to get into biology from a hi-tech successful experience is the lack of understanding of the development. Mac or IPhone or DeepBlue are the same, atom by atom from the point of its construction according to a rigid engineering plan. The human (or mice for that matter) go from a single cell to a fully formed and functional organism capable of very complex behaviour. Yes, it is all encoded as instructions in our genome but it is more like
    “All the human life’s a stage, and all the genes and siRNA merely players: they have their exits and their entrances; and one gene in his time plays many parts, his acts being seven ages…”
    Nothing even remotely comparable to what we have in even the most complex computer systems.

  37. John Galt III says:

    these guys do a good job with thought-provoking content. this is at least tangentially related to the discussion of innovation.
    http://www.thebaffler.com/salvos/of-flying-cars-and-the-declining-rate-of-profit

  38. John Galt III says:

    I missed whether someone already has pointed out that Thiel has a lot of visibility recently because he has been out touting his latest book. I’ve got a handful of links with interesting reading on the book and his thoughts.
    I have to say, what a fascinating character, full of contradictions. He’s probably a quick study, so likely will figure out that there are some differences between writing a website to shuffle money and thwarting death.
    Thiel, whose net worth is reported to be $2.2 billion, is Silicon Valley royalty, and a singular figure even in that rarefied world. He is a gay practising Christian, a libertarian who has thrown money and support behind the political campaigns of the Republican John McCain and the Libertarian Ron Paul, and who sits on the steering committee of the Bilderberg Group – the elite band of the rich and powerful from politics, industry and business that convenes each year to discuss nobody-outside-the-inner-circle-quite-knows-what. Above all, he is a man with a utopian belief in the power of technology to change the world.
    http://www.telegraph.co.uk/technology/11098971/Peter-Thiel-the-billionaire-tech-entrepreneur-on-a-mission-to-cheat-death.html

  39. This seems pertinent to my post (linked in my handle) on why drug design is not like airplane design – it’s the non-linearities, emergent phenomena and basic lack of information and predictive power
    One of the things that the public at large still does not appreciate about drug discovery is how hard it is to predict even very simple phenomena, like solubility and membrane permeability.

  40. thiel is right says:

    I agree with Thiel. Some people here criticize him saying that he’s an app developer or someone who created a website. It is not important what he did. His opinion matters and it is important. So, instead of logical fallacies let’s focus on what he said. We read here on this blog and on several other websites how pharma companies are trying to avoid taxes or going after smaller companies with promising projects instead of developing their own drugs. I also agree this is anti-technology. Let’s say I discovered a drug and it is approved. I am planning to sell it with some profit and this big company X comes and makes a generous offer. If I am a businessman, I’d sell my company (and my technology, drug etc.) and live a happy life. On the other hand, if I am a scientist and science is my only passion and responsibility, I’d reject the offer.
    I think Derek is wrong when he lists Apollo and Manhattan projects as engineering projects. Just because a huge rocket was made and carried people to outer space does not mean it is an engineering project. You should also consider the discoveries made throughout the project that helped the scientists design better rockets, fuel cells, engines, spacesuits, materials etc. People learned more about the properties and functions of their designs and compositions of materials. These led them to come up with better ones. We now have a group of scientists/astronauts live in space and does biochemistry experiments. There are cargo rockets that carry new experiments, goods, fresh food to these people regularly.
    Manhattan project also showed the importance of nuclear energy. It is not a coincidence that tens of nuclear reactors (small and large) were built more efficiently after the project. They learned how to dispose of waste better, they learned how to cool a reactor safer etc. People were working on isotope separation, but Manhattan Project I think made it quicker to find the answer. Nuclear submarines, airplane carriers are direct results of the project I believe.
    These projects are discovery projects rather than engineering ones. Just look at how many theoretical and experimental physicists, chemists, material scientists worked on Manhattan Project.
    People in drug discovery keep saying that how hard and expensive to develop a drug (I agree), but they almost always fail to realize that the ultimate goal of their companies is not drug discovery. The companies are managed by businessmen and the goal is to make more money. If another sector is more promising, I am sure they won’t hesitate a second to switch the course of the company.
    I might be wrong of course, but this is what I think.

  41. Bernard Munos says:

    What a great thread! The best one I’ve read in a while. Thank you all for articulating your thoughts so clearly.
    At the risk of simplifying a bit, the core of the argument seems to be that physics is a predictive science whereas biology is not. Whether that difference is inherent to the two sciences, and will remain so is an open question that I am afraid we cannot settle. Those who believe it is implicitly assume that biology is MUCH more complicated.
    I am not so sure. Physics was not always predictive, and had to deal with its share of enigmas (wave-particle duality, Schroedinger’s cat, etc). Quantum mechanics is not especially straightforward. It took us decades to understand particle physics — to the extent we do — with new “missing” particles painstakingly discovered as science progressed.
    What physicists did, however, that biologists should emulate, is that they have been much better at investing in basic science and eliminating knowledge gaps. There was a massive investment — much of which publicly funded — to understand physics at the particle level. Look at all the particle accelerators. We are not seeing that in biology. The Human Genome Project was completed in 2003, but there are scores of genes whose function remain unknown. There is an unknown number of pathways that are waiting to be discovered. Until recently, much of the DNA was thought to be “junk”, and we have barely started to figure out epigenetics. We need MORE investment to eliminate these knowledge gaps, but NIH’s budget is not keeping up — even shrinking in real terms.
    We also need a change in attitude. Physicists have been much better at joining hands to address these issues. Look at how freely they share insights and research findings on ArXiv! In contrast, until recently, precompetitive research in pharma was regarded as Utopian, and bitterly opposed by many industry leaders. Same leaders peferred to repurchase their companies’ shares, rather than make the sort of investments that would increase R&D productivity. Sematech was formed in 1987. Transcelerate was launched in 2012.
    Its not that money is lacking. Industry spends $140 billion in R&D each year, but rather than addressing problems that would make R&D more predictive, it prefers to treat each project as one-of-a-kind. Clinical research keeps failing because it rests on wrong hypotheses about diseases. Yet, until very recently, there was little interest in pooling knowledge and brains to come up with better hypotheses. Most companies were reluctant to disclose drug candidates failures, in the hope that their competitors would waste their money by repeating their mistakes.
    When biologists know as much about biology than physicists know about physics, biology will be more predictive, and our ability to build models that can simulate cell dynamics will be greatly enhanced. Perhaps we will never reach the level of predictability that physicists enjoy, but their is still a lot that we can do to get much better at what we do.

  42. JIA says:

    @40
    Mr.Munos, I agree with some of what you say — a large-scale concerted effort to fill in our gaps about the thousands of unknown genes and proteins may indeed be helpful, as the Human Genome Project was helpful. But I think the analogy with cyclotrons and particle physics is faulty. If a scientist at a US cyclotron discovers the mass of a sub-atomic particle, a scientist in Russia or Asia will get THE SAME RESULTS with “their” atoms. Properties of atoms don’t change. (Yes I acknowledge the difficulties of measurement, variations in equipment, etc — but the fundamental thing being measured does not change.) Now compare that to figuring out what gene X does. The answer is, it depends. Oocyte, zygote, embryo, different adult organs, under which enviromental conditions of growth factors, etc — you will get many many answers (that are all correct!) about “what does gene X do”. But you can only find out those answers one at a time. Do the properties of a subatomic particle change depending on what you were feeding its mother? The properties of gene in an organism can very much be dependent on factors like that. The term “emergent properties” is spot on for how biological systems work.
    Frankly, I think this science vs engineering debate is a sideline to the original point by Theil, which was about the anti-technology behaviour of large established companies. I agree with him that pharma is very often anti-technology. That doesn’t mean that a “tech approach” to biology will suddenly yield dozens of drugs.

  43. Carl Pham says:

    The basic physics underlying semiconductor chips — very little of which was developed by chip companies, by the way — is easy. Anyone who understands 1-D calculus can be taught the essence in a few weeks. Just because quantum mechanics is *weird* doesn’t mean it’s *difficult*. In fact, it’s generally easier than its classical counterpart. If you had to do Newtonian physics to understand chips, you’d have to work a lot harder.
    Same thing with the atomic bomb. The physics of fission are pretty easy, once you know it’s possible. Then you have to do a lot of measurements of cross sections and rates, and then you have the actual engineering problems of building an implosion to the unusually exacting standards of nuclear reactions (which means you need microsecond, not just millisecond, precision, because of the far greater speed of nuclear reactions). These are, as Lowe says, entirely engineering problems. Difficult, to be sure, but known to be soluble and the solution of which at least generally comes in a time that is inversely proportional to the money and people working on the project.
    Doing something like curing Alzheimer’s, where the basic mechanism is as yet almost all unknown — all there is is phenomenology, anecdotes, and epidemiology — is entirely different. First, you need one or more Big Insights or Experimental Breakthroughs, and the time for those to occur is generally unaffected by the number of (ordinary) people working on the problem, or the money thrown at it. You need, if not a Newton, at least a Pauli or Fermi, and those are sporadic uncontrollable events. You don’t get more of them by having more IQ 125 people major in biology in college, or having the NIH lower its funding standard so 50% of grant applications in Alzheimer’s get approved instead of 20% (or whatever it is). That’s like thinking you can build a rocket ship faster by hiring a lot of NFL linebackers and McDonald’s assistant managers. In this particular area, quantity is no substitute for quality.
    A much better analogy in physics for the problems of molecular biology are the strongly many-body problems, which share with biology the fact that their difficulty lies in the huge number of strong interactions, simple as each interaction may be.
    So…what progress has physics made on the problem of the self-energy of the electron in QED? How about the energy of the vacuum, and the cosmological constant? Those problems have been under study since the 60s, at least, and progress — if you’re honest — has been limited to ad-hockery and redefinitions, nothing the least bit satisfying. For that matter, can physics predict from first principles — no measurement data at all — the melting point of water? You’d think that be a trivial problem, given all the basic physics is known, and the only problem you have is applying it to 10^20 identical particles. But here we are, 50 years later, still struggling with it. That’s the nature of complex high-dimensional problems — of which molecular biology is one in spades. Progress is painfully slow and lurching, probably I would guess because the nature of the problem doesn’t fit well with our conscious rationality processes. We naturally think in chains and lines, with the occasional branch. These problems are all deeply net-like, with all the magic occuring via emergent phenomena. Human reasoning is terrible at dealing with emergent phenomena. (If it were not, we would be able to govern ourselves politically far better.)
    I’m certainly not saying chips or programming PayPal is *easy*. Neither are. They’re quite difficult. But the difficulty of, say, curing Alzheimer’s is that difficulty squared. Or perhaps raised to its own power. If you think achieving Moore’s Law represents the outer limit of human ability in terms of coping with challenges — you’d best just give up on conquering chronic disease. The only way you can believe in the latter is if you believe chips do *not* represent anywhere near the limit of human ability. Call them “very difficult” if you like, but then you need some much stronger adjective for what molecular biology has to achieve.

  44. I think everyone agrees that biology needs to spend more effort on basic knowledge-gap-filling activities. This is obvious.
    The problem arises in figuring out what those activities ought to be. Analogies to physics and chip design are facile and wrong. The problems are so vastly larger and more complex that simply scaling up the approaches used for physics and chips is probably not going to work.
    You probably need to hack around randomly collecting stray bits of knowledge until some sort of tractable methodology becomes clear. Then you have a methodology, an intellectual stance, a set of ideas, which you can scale up.
    Not being a biologist I don’t know where on this path the field is. From the outside it feels like ‘pretty early’
    Note also: there’s no law that says this stuff had to be humanly possible. It’s possible that there is no approach, there is no intellectual stance, there’s nothing that’s going to work. There’s no law that says that we don’t have to just blunder around in the dark forever.

  45. Morse says:

    @Bob Warfield
    “Folks, the fact that chips are human designed and organisms are not is a Red Herring. You’ve mismatched the components of the analogy. The chip is the drug”
    The burden of proof is on you in this case. You can make an analogy between just about anything, for example, a squirrel and a doorstop. It doesn’t mean the analogy is at all useful. And in this case the analogy is not because what’s expected of a chip and what’s expected of a drug are separated by orders of magnitude. If the standard for success in making a drug were equivalent to that of designing a chip then most drug programs would in fact be judged successful. Speaking from my own nearly 15 years of experience in the pharma and biotech industries (and no drug approvals), nearly every program I participated in reached a stage of designing a drug candidate that achieved its goals within an artificial system (e.g., desirable biophysical characteristics and activity in bioassays). Failures mostly occurred at later stages.
    At this point I don’t think you can even compare the magnitude of the task facing these two sectors. When the tech industry gets to the point that I can take my computer in and have its catastrophically failed hard drive and burnt out motherboard repaired then maybe we can take this talk seriously. And I mean repaired, not replaced. And you can’t open the computer, because this patient is afraid of surgery.

  46. Morse says:

    @Bob Warfield
    “Folks, the fact that chips are human designed and organisms are not is a Red Herring. You’ve mismatched the components of the analogy. The chip is the drug”
    The burden of proof is on you in this case. You can make an analogy between just about anything, for example, a squirrel and a doorstop. It doesn’t mean the analogy is at all useful. And in this case the analogy is not because what’s expected of a chip and what’s expected of a drug are separated by orders of magnitude. If the standard for success in making a drug were equivalent to that of designing a chip then most drug programs would in fact be judged successful. Speaking from my own nearly 15 years of experience in the pharma and biotech industries (and no drug approvals), nearly every program I participated in reached a stage of designing a drug candidate that achieved its goals within an artificial system (e.g., desirable biophysical characteristics and activity in bioassays). Failures mostly occurred at later stages.
    At this point I don’t think you can even compare the magnitude of the task facing these two sectors. When the tech industry gets to the point that I can take my computer in and have its catastrophically failed hard drive and burnt out motherboard repaired then maybe we can take this talk seriously. And I mean repaired, not replaced. And you can’t open the computer, because this patient is afraid of surgery.

  47. @ 43 says:

    I am following your logic now :
    Just by looking at the structures of most drugs, I can easily say that it is very easy to find a drug. They are mostly very simple molecules. In fact, Derek made a list of low molecular weight ones a few weeks ago.
    So maybe biologists, medicinal chemists, pharma companies are going after wrong targets and candidates instead of collaborating with scientists and being open about their research? Maybe they are wasting their time and resources?
    We can now send a rocket into the space and land on an asteroid. If this is not harder than drug discovery, I don’t know what is.

  48. John Novak says:

    Several points here as a practicing electrical engineer and research computer scientist:
    1) While it’s true that very few big projects are 100% engineering or 100% science, I still think Derek is right in classifying the Apollo Program and the Interstate Highway system as basically engineering programs with some research content. I might nudge the Manhattan Project a little closer to the science side, but not by much.
    2) With only a few exceptions, I’m not even sure what kind of metric the biological sciences would put together to come up with Moore’s Law-like curves of progress. Number of drugs per year? Sales per year? Lives saved per dollar per year? Time to market? None of those make sense in the way the nice pure GFlops/dollar or GB/penny numbers make sense.
    The exceptions are DNA analysis or synthesis base pair/dollar curves which *do* make sense and follow nice log curves– I think they’re calling them Carlson’s curves, these days.
    I’m sure Thiel would want to claim those for computer science, but, while there surely is a lot of computer science behind some of those methods, they’re still fundamentally impossible without the chemical techniques to enable them.
    3) Does Thiel even know how to *spell* “IRB” much less how to manage large scale trials with regulatory red tape? You fundamentally *cannot do* medical research in the same way as you do consumer grade, or even industry- or research- grade software development.
    I don’t even do medical research, and I know that, having suffered through a few small IRBs myself. Unless Thiel wants to address tha– and I’m speaking as someone who should be very sympathetic to him– the proper response is to just laugh him off the stage.
    (As a prominent libertarian, he might actually be willing to address it… but in fairness, he should be acknowledging the issue, at least.)

  49. Michael Byrnes says:

    The Economist’s Ryan Avent with a good rebuttal to Thiel’s general argument:
    http://www.economist.com/blogs/freeexchange/2014/09/innovation

  50. matt says:

    I’ve got some familiarity with the fabrication part of semiconductors, so I see where Bob Warfield is coming from (I’d say he’s working more on the Grove Fallacy than Thiel’s statement). But I agree with posters here, the fundamental complexity is many orders of magnitude more complex in biology. Compare the chemistry of a silicon etching or doping system, for example, with the chemistry of protein-protein interaction between two proteins that may have thousands of amino acids.
    I’d venture to say a single eukaryotic cell may be more complex than anything humans have ever designed. Banks of hydro generators, motorized roadways with automated tagged delivery systems, automated garbage collection, banks of “3-D printers” which can construct any part needed including themselves, sophisticated sensors and signaling systems, all of which are adjusting to present conditions and inputs. Some of you industry guys might say, forget the cell, you could just take flippin’ G proteins and GPCRs, and those are complex enough for humans to study for decades.
    We humans have trillions of cells. And these cells are all different, some subtly and some dramatically, and every one of mine are different from every one of yours, and that constant differentiation is a stochastic process which is a feature, not a bug. All of this even before considering collections of thousands/millions/billions of these working in concerted fashion to produce an organ like a kidney or an eye or a brain.
    If you think a little hard work will reduce all of this complexity to something like Newton’s Laws of Motion, I think you are deluding yourself. Rules of thumb which are generally and approximately true, yes. The precision of physics? Never.
    Fortunately, pharma doesn’t have to design this beast, but unfortunately it DOES have to understand it, at least enough to guess what’s going wrong and guess how a drug will interact with it.

  51. JJ says:

    I am not a chemist nor an engineer. But rather than debating whether designing chips are as challenging or not as challenging as discovering drugs, I would love to see Peter Thiel or other tech millionaires to start a pharma company. I hope they can find new antibiotics, cure for cancers, etc and I would be happy if they become more famous and immensely richer because of that.
    Please prove that the whole pharma industries are wrong and find more drugs for the benefit of humankind.
    Sadly, it is not happening yet.

  52. InsilicoConsulting says:

    @11 Bob,
    I wholeheartedly agree with your viewpoint. Most people here are naysayers who will crib and crib about how hard fundamental biology is and how hard drug discovery is.
    The point that fundamental physics, material science, applied quantum physics etc is very hard for it to make electronics possible does not make it beyond their BBB.
    Having a complex system with many parameters e.g. climate, human beings certainly makes predictions or a systems analysis hard to do. But so does coming up with fundamental new ways of thinking in physics which can lead to breakthroughs at an applied level. The fact it took many geniuses people like feynman, freeman dyson and many others to achieve something is always forgotten e.g. manhattan proj.
    The manhattan project was a mix of cutting edge applied science and engineering, not only engineering is just not clear to most respondents here.
    All the same, most of these will never be the ones to work on cutting edge concepts. They will happily question whether targeting beta-secretase works. Well, find out intelligent low cost ways to find out how to measure whether it works! You are the experts and you should be the disruptors!
    The main problem is that these commentators here confuse many moving parts in a dynamic machine and knowing their relationships kind of complexity
    Vs
    Complexity inherent in non-intuitive , extremely mathematically challenging (e.g. symmetry), revolutionary concepts in physics.
    Physics for them is simpler than biology as it has 4 simple equations of electromagnetics for example. Most will be challenged to come up with something equivalent in biology with even 1/10 th of the complexity even for a single pathway.

  53. Maple says:

    It may be useful to review examples from the history of drug discovery – for instance the development of acid blockers for peptic ulcer disease leading to a Nobel prize (in part) for Douglas Black. Whilst the industry refined and significant improved the approach with the development of PPIs it was recognition (many years after the initial launch of cimetidine) that peptic ulcer disease was in fact the result of infection with H Pylori and that antibacterial treatment offered a cure (work that also resulted in a Nobel prize). That recognition essentially came from a set of phenotypic observations resulting in a human challenge study which replicated the disease.
    The pathogenesis of Alzheimers is likely to be infinitely more complex- at the time of diagnosis the underlying disease process may have been developing for 20-30 years (or more) and the observed pathology is the result a series of pathological changes (neuronal death, repair and inflammation) which are likely to be irreversible using currently available interventions (even assuming we understand them which we mostly do not).
    In the absence of disease-relevant phenotypic screens which accurately represent the evolution of the disease we have no opportunity to intervene. Assuming we had phenotypic screens, the best hope would be primary prevention in susceptible patient populations with relevant biomarkers which accurately reflect disease progression (and treatment responses). Some of these may be in place already – genetics identify disease sub-populations and PET MRI techniques represent end organ pathology in more advanced disease. Joining up the dots and providing credible phenotypic screen for early disease may provide an outside shot at developing an effective medicine for primary prevention. But figure the clinical trial implications for primary prevention – assuming you could identify highly susceptible patient population how long would it take to get a new medicine through regulatory hurdles and who would invest in this ?
    Regulatory authorities would have to recognise early biomarkers (blood and imaging)for purposes of regulatory approval – how likely would this be in our ultra risk averse regulatory environment ?
    It is ironic that we might already have small molecules that could ameliorate disease progression (the cimetidine analogue) but we are a long way from finding out what these are and how and when to intervene…..

  54. David Borhani says:

    @53, Maple: m’ thinks you meant to write Sir James Black, the inventor of Cimetidine (and Propranolol), and Barry Marshall, the discoverer (via self-ingestion) of the ulcer-promoting nature of Helicobacter pylori.

  55. aa3 says:

    The Moore’s law for biotech should be average human lifespan. Kurzweil and probably others observation is that in the 1700’s we were adding something like a few days to the average life expectancy each year. Then it rose to a few weeks, and today it is in the months being added each year.
    Just as in computing it always appears there is walls preventing further progress not far away. Like the heat wall around the year 2003 in computing.
    And progress can come from way outside the box. Like in the human life expectancy during the 1700’s and 1800’s a lot of the rise came through increasing the protein content in grains. With much more protein, the people were more durable for surviving infectious diseases.

  56. grade baa says:

    Why listen to this guy at all? As far as I can tell he did one thing successfully – found PayPal and then manage to merge (not buy) X.com and gain Elon Musk in the process.
    Everything else has been investor, not inventor based – his Clarium Capital fund that was betting on hyperinflation, his founders fund that bets on internet companies, etc…
    Other than having money I see little actual knowledge or experience behind his views. Maybe a step up from listening to Paris Hilton, who is also rich, but not much.
    At least Andy Grove ran Intel for a couple of decades

  57. grade baa says:

    Why listen to this guy at all? As far as I can tell he did one thing successfully – found PayPal and then manage to merge (not buy) X.com and gain Elon Musk in the process.
    Everything else has been investor, not inventor based – his Clarium Capital fund that was betting on hyperinflation, his founders fund that bets on internet companies, etc…
    Other than having money I see little actual knowledge or experience behind his views. Maybe a step up from listening to Paris Hilton, who is also rich, but not much.
    At least Andy Grove ran Intel for a couple of decades

  58. Chris says:

    Like most here I defend the pharma “process” as not being anti-technology. But the larger community (where Thiele lives) does see a couple of things differently – unaligned incentives for antimicrobials sees some potential ‘cures” left relatively under-researched. And the crazy high cost of certain new cancer treatments leaves people feeling the pharma companies “could” fix some diseases if they thought they could also justify such large sums. As an insider, the innovation stagnation came around the same time as best practices were imported into drug discovery teams by the MBA-ers. Pedro Cuatrecasas wrote about this early.

Comments are closed.