Skip to main content

Drug Development

Open-Source Pharmaceutical Babble

The topic of “open-source” drug discovery is an interesting (and potentially important) one. It just keeps coming up, but one of the problems with it is that it presents a terrible opportunity for vagueness. Too much of what I’ve read on the subject is hand-waving.
I’m afraid that the key parts of this column fall into the same category. It’s by Jackie Hunter, formerly of GlaxoSmithKline. The lead-up parts of the piece are fine, where she lays out some of the problems facing the industry. But then we get this vision:

In the future, the most effective pharmaceutical companies will be hubs at the center of a network of collaborators and suppliers, focusing internally on their core competencies, which might include medicinal chemistry, execution of clinical trials, or sales and marketing. They will facilitate interactions across their network to stimulate the development of innovation ecosystems.
The resulting opportunities to expand beyond traditional products and markets will enable pharmaceutical companies to evolve into companies that offer a range of health-care solutions. These will include not only prescription medicines, but also diagnostics, branded generics, and technologies that support personalized medicine, as well as so-called “neutraceuticals” and other “wellness options.”

And that’s it; that’s the payoff. We’ll all just hop to it, enabling and facilitating, expanding and evolving, stimulating and focusing. None of those are concrete verbs suggesting real courses of action. Whenever you see someone slip into that sort of talk, you can be sure that (at the very least) they have difficulty communicating whatever specific ideas they have. Or (more likely) that they don’t have any specific ideas to tell you about at all.
Not that I can blame Jackie Hunter. I don’t have a lot of good suggestions at the moment, either. But if you read that column closely, it says (on the one hand) that the problems of the industry are so large that single drug companies probably can’t deal with them. Fine. Then it goes on to say that dealing with them will probably reduce the size of drug company R&D organizations. The connection between those two ideas is presumably hidden in that ball of fuzz I quoted above.

36 comments on “Open-Source Pharmaceutical Babble”

  1. RB Woodweird says:

    She is talking about the legendary missing Step 3:

    Step 3: ???
    Step 4: Profit!
    (Also, The Legendary Missing Step 3 would make a good name for a band or for a book of management techniques.)

  2. PharmaHeretic says:

    What were you expecting.. This is standard MBA-speak.
    Only fewer PHBs and more Dilberts can save pharma, big or small.

  3. MTK says:

    Even assuming it isn’t just babble, but a real strategy toward “a range of health-care options” that include diagnostics, branded generics, etc., isn’t that what was tried in the late 80’s and early 90’s? Didn’t Merck buy Medco? Didn’t SmithKline Beecham have SB Clin Labs for diagnostics? Isn’t what she mentioned basically what Abbott and J&J are to a large degree?
    I guess what I’m saying is “Been there, done that.”

  4. HelicalZz says:

    There has been plenty of commentary around how pharmaceutical business models need to change and adapt. Sure, adding new revenue streams where practical is always useful, but spending time developing models that don’t have clear revenue paths is a fools errand. Industry commentaries like Progressions (Pharma 3.0) have some practical comments, but are also loaded with vacuous unclear business model suggestions.$FILE/Progressions_2010_Media%20copy.pdf
    Pharmas key business will continue to be selling medications. How they get to and interact with the consumer is and will change, but that is the key revenue stream. It does seem that the company will have less control over the use of their products and the positioning of them than they have in the past, which will be a large challenge.

  5. K says:

    All meaningless buzz-words, zero hard and fast content. That could have been written (pretty much word-for-word) by any middle/senior manager within GSK

  6. processchemist says:

    Uhm… model already implemented:
    – inner R&D substantial cuts
    – using CMO (Asian)
    – using CRO (Asian)
    – using Biotechs (western)
    Total result: a loss of inner technological capabilities.
    Talking about my line of work, a couple of years ago I was discussing with a project manager that insisted to have a scaled up process using anhydrous MIBK from distillation over phosphoric anhydride. No use talking about azeotropic drying with KF control. The joys of this brave new world.

  7. LAM says:

    There are two parts to this. One is a specific descriptive model in evolution from current small molecule Pharma to a more diversified company. As MTK noted, not unlike J&J. & some efforts by other Pharma having been much less successful.
    On the other hand, I totally agree with her sentiments that the environment of small drug R&D will require more downsizing and consolidation. The challenges continue to grow for delivery of new drugs. It will continue to hit individual companies when major patents expire and profits can’t be replaced by internal growth, but are not restricted to this problem alone: increased regulatory presence, more safety scrutiny after launch, increased concerns of cost, more evaluation between potential drugs vs placebo, larger & more costly clinical studies, lack of true validation of biomarkers to valued end-points, are only a few of these increased challenges.
    Knowing JH for some time, these statements are not a large surprise: great for visionary statements, presentations which can look good to some Pharma management struggling to find new direction, but much less impressive if one looks at delivery, implementation, & business success (eg what new drugs or business profits?) What is she selling now, in life-after-big Pharma?
    Overall, Derek, you are only giving JH more of a platform for her to get her Jones, need for attention. I don’t know why you provide even this amount of credibility.

  8. RM says:

    Before I read the full article, I would have said I understood the quoted section, or at least the first paragraph.
    My interpretation is that instead of traditional monolithic companies, which see a drug through from the initial research stages all the way through to marketing, in the future Pharma is going to rely more on outsourcing. Instead of being a “jack of all trades”, each company will have a particular focus (a “core competency”) – they’ll be a research co., a development co., an animal model co., a clinical trial co., a marketing co., etc., and a drug will get passed between them in loose and fluid partnerships. We sort of see this today with startups specializing in discovering new drugs which will get passed on to a big pharma for subsequent steps, and the contract clinical trial companies.
    The second paragraph gets fuzzier, and I would only assume that it means that, freed from the hassle of developing drugs themselves, the marketing companies can market a range of healthcare products, not just traditional pharmaceuticals. (I can only assume she’s ignoring the other companies in the pipeline, as a small drug discovery company isn’t going to have much success working with “branded generics”.)
    That’s how I interpreted it, at least until I read the full article, and encountered a wall of MBA jargon and technobabble. There may be ideas about the future of Pharma in that article, but it’s just as likely it’s a Rorschach blot, with each person projecting their own biases upon it.
    (BTW I still don’t see how any of this is “open source” – there’s more to open source than “be more collaborative”.)

  9. vvhome says:

    Derek, Your are absolutely correct. Open-source drug discovery seems to conjure up Utopian dreams of all these moving parts happily and harmoniously coming together for the greater common good. What worked for IT will not necessarily work for pharma. Some specific projects, particularly those on diseases like AIDS, malaria, leprosy, TB, etc., might get some traction though.

  10. DLIB says:


  11. emjeff says:

    Did Jackie get paid to write this nonsense? I’m in the wrong line of work…

  12. daen says:

    First of all, the article is less than 900 words; how much detail do you expect? Secondly, she explicitly closes the article by acknowledging that creating a network of focused-competence organizations is not necessarily the same thing as open sourcing everything. Thirdly, the open source movement is comparatively new. I remember in the early-90s when IBM declared a loss of more than $8bn. Since then, it has rebuilt itself on the open source movement. However, it has topped the list of patent registrations for the 17th year, while pledging not to enforce a number of keys patents relating to the open source movement. And remember, many of IBM’s patents relate to research undertaken in some very tricky areas of physics. Now, nobody in their right mind in the late 80s, before IBM started declaring its series of billion (and later multi-billion) dollar losses would have predicted this. And yet, here we are today, with the company’s 2009 net income at $12.3bn. Now, in my simplistic view of the world, if one research-based industrial giant, previously viewed universally as arch-conservative (reactionary, even), can perform a radical and yet profitable about-turn in the face of major crisis, then is it really stretching credulity to believe it can be done elsewhere, in biotech and pharma, as JH suggests? As for “Utopian dreams of all these moving parts happily and harmoniously coming together for the greater common good”, vvhome has conveniently forgotten, or is ignorant of, how IBM defended Linux from SCO’s lawsuit in 2003. Don’t be so naive: open source does not mean holding hands and tripping out in a field while dressed in tie-dyed T-shirts. It’s a viable and valid business model, with acknowledged intellectual property which is both protected and shared. The difference from pre-existing business models is that there is now a broader and more complex set of reasons for doing so, the profit motive being just one (but still a powerful one).

  13. Boghog says:

    From my reading of the linked column, the main thesis is that open source in the pharmaceutical industry would take the form of pre-competitive collaboration. This in fact is not new, since there previously have been a number of industry consortiums. What may be new is an increased importance of these consortiums and a focus on the main causes of clinical failures, lack of efficacy in humans and unexpected toxicity. The proposed solutions would be consortia addressing target validation and biomarkers for efficacy and toxicity.

  14. Hap says:

    Is it clear how open-source will work, though, where 1) there is massive infrastructural and other costs to be shared among the participants and 2) where the physical product (and its costs of delivery) is a non-negligible part of the overall cost? IBM advertises itself as a sevices company. Pharma doesn’t sell services, it sells products, without which it doesn’t have anything. There are also people between pharma and the consumer (doctors) who already provide the services which someone might hope to sell, unless the knowledge of drugs and their effects is kept tightly by pharma, which wouldn’t appear to be a benefit to anyone but pharma.
    The question of whether open-source pharma can work doesn’t explain why the concept can’t be explained clearly. If you have only 900 words (probably more words than most newspaper articles), spending half of them on management buzzwords doesn’t help.

  15. ronathan richardson says:

    I heard Mark Murcko of Vertex fame give a talk recently in which he decided to put in a plug for open-source drug discovery as one of the keys to healing pharma. From what I could glean, his definition of open source was that people who discover drugs but aren’t in a position to develop them should give me their IP, mmnow.

  16. processchemist says:

    I think that the main problem is the foggy “pre competitive” term.
    The birth of OPRD is an example of pre competitive cooperation: process chemists (and companies) that found valuable sharing experiences to not be forced to “reinvent the wheel” everyday.
    What’s pre-competitive in drug discovery?
    specific knowledge of biological pathways? Clinical data of failed trials?
    I was thinking (and maybe I’m wrong) that in industrial drug discovery both things are keys of
    the competition.
    In the software arena between the operating system and the end user solution there’s a value added field. So an open source OS or DBMS can be a great resource, if you focus in implementation of solutions (but Microsoft keep selling OS…).
    Another point: to work on a Linux module you need a PC (few bucks), a linux distribution (free), and your developing skills.
    Not exactly the order of magnitude of investments needed by drug discovery. Who’s gonna fund this “pre-competitive” domain? The suspect, as usual, is that in the model big pharma take the results while the tax payer funds the pre-competitive activities. An hypotetical true open source for pharma should be a model where pre competitive knowledge is shared in the same way funds and resources are shared. Not exactly the current deriskify and diversify.

  17. LAM says:

    Comparisons have been made for many years between making new drugs and developing new chips and software. Pharma has tried to revamp themselves using revised goals in areas like discovery or cycle time to mimic the expectations from the world of making new generations of chips. But, the businesses, regulations, safety testing requirements, need to demonstrate efficacy in long term studies simply don’t translate to the world of computers and software. If Apple makes a mistake on a new i-Phone, all they need to do is provide a plastic cover at a few cents per item in a cosmetic attempt to make it all better. If a big Pharmaceutical company makes a mistake in evaluation of safety for a new drug, it can cause health problems to many, cost millions or billions in liability claims, and can put the company at risk for their long term survival. I’ve never seen how the comparison works, and even more so in today’s world after events like the recent FDA review on Avastin which should never have been approved for use in breast cancer without survival data, and the “no new news” re-review of Avandia. When chips have to go under this amount of review and scrutiny, then the comparisons become more equivalent.

  18. Derek F says:

    I like the former “Head of Science Environment Development” – pretty easy to see that she comes from the woo end of the business.
    Also “neutraceuticals” – how many people has GSK neutered in the past few years?

  19. daen says:

    @Hap: Well, infrastructure costs and safety issues (as pointed out by LAM) are already problematic under the current model. Product versus service is more interesting. Of course the differences between what IBM did then, and what big pharma would have to do now, are significant. The similarities, however, intrigue me – except in this case, it’s an entire industry running into the end-buffers rather than a single company. What I also find fascinating is how much agreement there is across the industry that “something has to be done”, and yet so much resistance to trying something radical. Mergers, acquisitions and restructuring don’t count – and don’t seem to work very well, as Derek has commented on many, many times. So, while it’s true that the Devil is in the details, open source has been an unexpected, crisis-breaking and profitable approach in one high-tech area so far, and I think that JH is quite justified in pointing out that it might possibly work for pharma, too.

  20. Trajan says:

    Interesting little insight into the Big Pharma mindset. Lots of buzz words, no substance except the usual ass gas we’ve come to expect from some one with her background.

  21. Hap says:

    The problem, though, with the open-source alternative is that it hasn’t solved those problems (because it hasn’t had them) and adds new ones. How do we split the expenses and revenue (or output) more generally from such ventures? (Incubators? Fund big infrastructure with lots of small groups and charge as rent and a piece of the pie if something good comes out?) Where does the money come from to set up the infrastructure? Even with areas where it’s obvious cooperation could help everyone (for example answering questions like “why do drugs fail trials?”, “what biomarkers work?”, and “how are drugs handled and metabolized by people?”), they aren’t done, so it seems like an even harder tack to assume that those problems and more can be solved when your business depends on it. People take daring and uncomfortable options options either when there’s no other choice, or when they think that choice can actually work. (We’ll discard the “Drugs will appear magically if we lay off R+D and keep the stock price high.” choice.)
    In addition, pharma already has people (stockholders) who are in the business specifically to make money. Adopting an ethic which bumps that down on the list of priorities probably isn’t going to go well. More generally, people are willing to share when they can afford to and when it’s relatively easy, and sometimes in other circumstances. With no alternative jobs, and no other support, I’m not thinking that being an open-source chemist is going to work well.
    I think that pharma is loath to open-source because ultimately it is interested in making money, first and foremost. Even if a model existed with relatively easy analogy to pharma where open source worked, its heart is antithetical to what has made pharma (and lots of other businesses, likely the ones from which their management came) successful in the past. People with less to lose in the current system, if anyone, are probably the only way that open-source pharma will be successful.

  22. daen says:

    @Hap: You’re right about not having to solve the problems – and that’s the challenging bit, for sure. As to divvying up the proceeds, there are always a large number of creative ways of doing that.
    I think you’ve hit the nail on the head: people take the path less traveled only when the other way is leading you into the forest full of bears/wasps/Smurfs. I am almost certain that in the closing years of the 1980s, IBM management would not have predicted that within 20 years a huge chunk of revenue would come from a business model that was not only anathema to them at the time, but was one which no-one could yet figure out how to profit from. Until they did. Only a multi-billion dollar crisis drove IBM to try something mould-breaking. Perhaps that is what it will take for pharma, too.
    Because if pharma companies and their stockholders really are profit-driven, having the life-blood haemorrhage year after year from your organization because of the existing flawed business-as-usual model, with no immediate recourse to stem the flood other than cutbacks and mergers, with the inevitable retardation of the R&D effort leading ultimately to even larger losses, does kind of narrow your options until your only choice is between forced adoption of something new and radical, or extinction.

  23. Theclifster says:

    ok Derek I am not sure if I agree or disagree.
    I do think that currenty large amounts of valuable data lies in redundant silos within a few organisations and that this should be
    a) Shared, more brains = more ideas
    b) Successful companies will be the ones who quickly recognize a novel idea and partner quickly with those who have it.
    Is this open source, no its not a shared collaboration of a source dataset, but it is open in the sense of you can see who is working on what, buying into what and partnering with whom.

  24. LAM says:

    HAP: I don’t get your logic. Are you trying to say that IBM management, employees, stockholders were not motivated and driven by sales, market share and profits?
    What IS different is that the technologies supported by IBM involve hardware & software infrastructures, chips, processors, bits and pieces of IT gizmos, much of which involving plug and play, or interchangeable parts or systems.
    I have not seen such easy modular processing in drug R&D. In fact, just the opposite. Biology is complicated, messy, often unpredictable, with still huge missing gaps in our basic understandings.
    But JH would like to let us think one can make drug R&D so simple and modular. It did not work for her at SB when head of a biology dept, in GSK when being a CEDD head, and won’t work now in her new business management gig (although she might be able to make some money from some poor unsuspecting clients).
    There are a lot of smart, clever, insighful, risk-taking people in drug R&D. To the point that if Pharma R&D were so modular, then today’s industry’s dilemas would already have been resolved, and we would already be within a new world order for delivery of new drugs.
    But, it ain’t so; it does not translate.

  25. dvrvm says:

    The problem I have with the splitting-up of the monoliths into smaller autonomous units is that from my point of view, the clinical development is just way too cost-intensive, which is from my point of view what drives the monolith-formation. If you’re rich enough to do the clinical trials, you might as well run the preclinical parts yourself and get a longer value chain.

  26. Joel West says:

    One problem is that there’s nothing “open source” about this approach. It isn’t even “open source biology.”
    Instead, it’s more of an “open innovation” approach.
    Sheesh. You’d think a CEO would get her buzzwords right. But then she has a PhD in Psychology and not an MBA.

  27. processchemist says:

    By the way, about IBM: the restructuration of the business model destroyed a mass of good and innovative technologies: the OpenDoc platform, OS/2 Warp, powerpc. And if you’re able to compare an old Thinkpad with a new Lenovo notebook you can *touch* the difference. The only winner left on the field is Microsoft with his totally closed source model, a de facto monopolist in the global PC operating systems and office applications arena. And they’re not exactly delivering excellent products (they never did).

  28. LeeH says:

    The quote gets a 2.6 out of 10 on Deloitte and Touche’s Bullfighter software.
    The whole article, on the other hand, gets a 4.1. Here’s the evaluation, as reported by the software:
    Diagnosis: You like to hear yourself write. Despairing of the thought of bringing a sentence to a close with something as demeaningly ordinary as a simple period, you shower readers with gratuitous, interminable and often weighty if not impossibly labyrinthine prose. Meaning lingers, albeit awash in a thick tide of metaphor and exposition that threatens to drown the writer’s message. Seek help.

  29. daen says:

    @processchemist: As far as I’m concerned, the last good version of OS/2 was 1.3. It was robust, fast and way ahead of its time (TRAP 000D allowed you to invoke the JIT debugger, replete with stack trace a la VMS, and that saved me many, many hours of debugging). I recently found out that a suite of my OS/2 1.3 server applications written in 1993 had been running in a bank in Berlin until 2007. OS/2 2.0 and Warp, on the other hand, were always playing catch-up with Windows emulation, and suffered for it (and Talgent was never, ever going to work in the IBM climate of the time). PowerPC still exists. And OpenDoc? That’s Apple’s OLE equivalent … Do you mean something else?

  30. processchemist says:

    In my experince best version was OS/2 Warp 4 (serviced). With A SCSI hard disk and 128 Mb of ram far faster than NT with the same configuration. Superior TCP/IP support, good posix compatibility with the emx package, and so on. Powerpc is confined to embedded systems, since Apple choose intel processors for the macs. For OpenDoc (the platform) you can have a look to wikipedia (it was an alternative to OLE, and completely different as concept – a common platform to incorporate any kind of objects) Also SOM was great: I had some fun writing small utilities for the workplace shell.

  31. because health matters says:

    “One can see the free software movement as a precusor for a “free hardware” or “free wetware” movement–one that will provide free libraries of designs for biological or nanotechnological products that replicators can be programmed to churn out.” Charles Stross (2002)

  32. Anonymous says:

    Roche is the place to be!! TRUST ME…

  33. davesnyd says:

    What’s there to understand here? My read of the Google search results for “Pharmivation” make it sound like she left GSK earlier this year, established a consulting firm with a focus on helping pharma companies do this open whatever it is she’s suggesting, and has now written a white paper suggesting they do it.
    Having said that, I think that @processchemist is right– this is largely a repackaged outsourcing and public/private collaboration to reduce the risk and investment by the pharmaceutical companies. If so, then I think the article’s author is reading the direction of the industry more or less correctly– more outsourcing, less internal effort and investment, and (the CEOs and boards think) higher and more reliable profits.

  34. Anonymous says:

    So in short, pharma companies will become like Enron?

  35. Having read this I thought it was rather informative. I appreciate you finding the time and energy to put this article together. I once again find myself spending a lot of time both reading and posting comments. But so what, it was still worth it!

  36. Eman says:

    Pharma’s…where my money is always at…the smartest business models and safe investments as they keep adapting not to only survive but thrive 🙂

Comments are closed.