Skip to main content

Chemical News

What Medicinal Chemists Really Make

Chemists who don’t (or don’t yet) work in drug discovery often wonder just what sort of chemistry we do over here. There are a lot of jokes about methyl-ethyl-butyl-futile, which have a bit of an edge to them for people just coming out of a big-deal total synthesis group in academia. They wonder if they’re really setting themselves up for a yawn-inducing lab career of Suzuki couplings and amide formation, gradually becoming leery of anything that takes more than three steps to make.
Well, now there’s some hard data on that topic. The authors took the combined publication output from their company, Pfizer, and GSK, as published in the Journal of Medicinal Chemistry, Bioorganic Med Chem Letters and Bioorganic and Medicinal Chemistry, starting in 2008. And they analyzed this set for what kinds of reactions were used, how long the synthetic routes were, and what kinds of compounds were produced. Their motivation?

. . .discussions with other chemists have revealed that many of our drug discovery colleagues outside the synthetic community perceive our syntheses to consist of typically six steps, predominantly composed of amine deprotections to facilitate amide formation reactions and Suzuki couplings to produce biaryl derivatives. These “typical” syntheses invariably result in large, flat, achiral derivatives destined for screening cascades. We believed these statements to be misconceptions, or at the very least exaggerations, but noted there was little if any hard evidence in the literature to support our case.

Six steps? You must really want those compounds, eh? At any rate, their data set ended up with about 7300 reactions and about 3600 compounds. And some clear trends showed up. For example, nearly half the reactions involved forming carbon-heteroatom bonds, with half of those (22% of the total) being acylations. mostly amide formation. But only about one tenth of the reactions were C-C bond-forming steps (40% of those were Suzuki-style couplings and 18% were Sonogoshira reactions). One-fifth were protecting group manipulations (almost entirely on COOH and amine groups), and eight per cent were heterocycle formation, and everything else was well down into the single digits.
There are some interesting trends in those other reactions, though. Reduction reactions are much more common than oxidations – the frequency of nitro-to-amine reductions is one factor behind that, followed by other groups down to amines (few of these are typically run in the other direction). Among those oxidations, alcohol-to-aldehyde is the favorite. Outside of changes in reduction state, alcohol-to-halide is the single most favorite functional group transformation, followed by acid to acid chloride, both of which make sense from their reactivity in later steps.
Overall, the single biggest reaction is. . .N-acylation to an amide. So that part of the stereotype is true. At the bottom of the list, with only one reaction apiece, were N-alkylation of an aniline, benzylic/allylic oxidation, and alkene oxidation. Sulfonation, nitration, and the Heck reaction were just barely represented as well.
Analyzing the compounds instead of the reactions, they found that 99% of the compounds contained at least one aromatic ring (with almost 40% showing an aryl-aryl linkage) and over half have an amide, which totals aren’t going to do much to dispel the stereotypes, either. The most popular heteroaromatic ring is pyridine, followed by pyrimidine and then the most popular of the five-membered ones, pyrazole. 43% have an aliphatic amine, which I can well believe (in fact, I’m surprised that it’s not even higher). Most of those are tertiary amines, and the most-represented of those are pyrrolidines, followed closely by piperazines.
In other functionality, about a third of the compounds have at least one fluorine atom in them, and 30% have an aryl chloride. In contrast to the amides, there are only about 10% of the compounds with sulfonamides. 35% have an aryl ether (mostly methoxy), 10% have an aliphatic alcohol (versus only 5% with a phenol). The least-represented functional groups (of the ones that show up at all!) are carbonate, sulfoxide, alkyl chloride, and aryl nitro, followed by amidines and thiols. There’s not a single alkyl bromide or aliphatic nitro in the bunch.
The last part of the paper looks at synthetic complexity. About 3000 of the compounds were part of traceable synthetic schemes, and most of these were 3 and 4 steps long. (The distribution has a pretty long tail, though, going out past 10 steps). Molecular weights tend to peak at between 350 and 550, and clogP peaks at around 3.5 to 5. These all sound pretty plausible to me.
Now that we’ve got a reasonable med-chem snapshot, though, what does it tell us? I’m going to use a whole different post to go into that, but I think that my take-away was that, for the most part, we have a pretty accurate mental picture of the sorts of compounds we make. But is that a good picture, or not?

24 comments on “What Medicinal Chemists Really Make”

  1. Duke says:

    This is a clear message to all chemists to add “Sonogoshira”, “Suzuki”, and “amide formation” to their resumes.
    Your friendly bubble headed HR person will now provide you with an interview.

  2. MTK says:

    Uhhh, I don’t think that would be a wise strategy.

  3. Anonymous says:

    “Now that we’ve got a reasonable med-chem snapshot, though, what does it tell us?”
    That it’s much easier to outsource amide bond reactions as opposed to C-C bond forming reactions? Surprise, suprise.
    I’d love to see the same analysis for those companies in the years before their mass outsourcing began (well before 2008). I think we’d have a different snapshot to look at but maybe I’m wrong.

  4. Sundowner says:

    Ah, this is not going to make more popular, but…
    In fact, my opinion about that the usual medicinal chemistry is that it is usually boring and very predictable. Honestly, to me, being a hardcore synthetic chemistry (NOT involved in total synthesis campaigns which look like the Napoleonic Wars, but methodology, haha), the usual J. Med. Chem. paper contains no interesting chemistry at all. The structures are interesting, the results are very useful… but the chemistry itself, I am afraid is boring: A, B, C and more of the same.
    So I am not surprised at all to know that a) that perception is there and b) that the raw data supported the ‘misperception’.
    And now, talking about my own experience, I have to say that many med chems look the synthesis that way. Without giving away many details (sometimes this is like working in the CIA or something), very recently we were working in the synthesis of a standard for a very big hit finding project, along other companies. One of the med chems wanted us to prepare a compound, chiral, long synthesis (15 steps), and as usual no valuable info was given in the patent (i.e., no chiral HPLC, no alpha… nothing). Well, we needed a lot of work to a) reproduce the synthesis b) check what isomer we were looking for and c) scaling up and improving the synthesis to get the grams they needed for the hit to lead campaign. To summarize, while other med chems were screaming and complaining about how difficult the preparation of that standard was, other groups prepared flat, easy molecules in half the time (4 or 5 standards each). It was quite depressing. In the end we succeeded and it turned out that the standard we had been given was the best pharmacological tool.
    So yes, many med chems have become… hmmm, maybe lazy. If you can do it in five or six steps is OK. If it is longer, ah, forget it. If it is chiral, then is ‘wa-do-ya-think-ar-ya-doin?’. No challenges, please.
    No guts, no glory !

  5. JV says:

    Your analysis recalls me a joke about a guy who was looking for a key where light was..
    Well, the true is that majority of medicinal chemists in indusctry do not publish at all. They do so only if:
    1) the project was closed
    2) they want to hurt competition (usually patent priority is involved)
    3) they want to confuse competition
    Published papers are a small fraction of overall operation, make things worse, which does not reflect reality in medchem research

  6. Nick K says:

    I second JV’s post at #5. Sampling error is a potential problem if we only look at publications in the open literature. Examining the patents produced by these companies would probably furnish data more representative of the actual lab notebooks. It would be fascinating to see how the style of chemistry has changed over the years. Like #3, I suspect that it would show that the chemistry has declined in innovation and complexity in recent years.

  7. patentgeek says:

    A perspective from 20+ years in med chem (since gone over to the dark side of patent law): I think the notion that some synthetic gunslingers have about where the intellect in pharma chemistry is applied is mistaken.
    Someone (it may have been Bob Silverman, but I could be wrong) once said “The problem is most usually what to make, not how to make it.” Crafting a candidate molecule, with acceptable bioavailability, toxicology profile, and efficacy, is an exercise in constraint and minimalism. You only have so many MW and logP units to spend to get the right mojo. It generally takes many variations to get the right one, and the quicker you knock out the analogues, the better your chance of finding a goodie in your lifetime.
    I came out of alkaloid total synthesis, and never viewed myself as paid in pharma to do fancy chemistry. The intellectual exercises that drove me were things like staring at three years of data, figuring out it consisted of multiple overlaid SARs, and then peeling back those layers to structurally separate the activities and optimize the desired one. Actually making the compounds was anticlimactic. (Seeing the data that confirmed my hunch was not!)
    The best compound I ever made – still in the clinic – took three steps to make and could be made by a high school kid. I didn’t view that as a negative, and neither did our process and clinical groups.

  8. RTW says:

    Well – from my experience this easily displays the effects of bean counter mentality of counpound counting, and reaction counting as a measure of lab success of the folks that work at the bench. Younger MBA group managers – thought that numbers where more important than answering fundamental questions about the leads that HTS where providing. So it degenrated into an exercise of adding only commercially available spaghetti to typically easily attainable cores.
    People that wanted to answer the hard questions by moving heteroatoms around, add small appendages, to the core, where usually slammed if it couldn’t be done quickly. Typically given one chance. This also adversly effect compound numbers, reaction counts remained the same or higher but resulting in more failure due to the risks involved. I saw instances where people made things they could buy to get their reaction counts up. And submit compounds that could never be tested (reactive or otherwise) to get their compound numbers up. So its no wonder we see this sort of results in a metadata analysis.
    Next question is in this analysis – how many compounds made it to phase II or beyond, given that the work done in these papers is actually years old before published?
    There is much more to doing bench medicinal chemistry synthesis, than the methyl, ethyl, butyl futile, unless thats all your management will let you do. Afterall they know better becuase they are managers, not actually working at the bench any more.
    Process development chemistry is often very interesting, in trying to optimize synthesis of compounds. Remember for clinical candidates the overall isolated yields for each step should be very high, and the isolated purity also quite high with any contaminates fully characterized. A good discovery chemist learns to take solvent, reaction, selection and purification methods into consideration when developing promissing leads.
    So – perhaps from the outside it does look rather dull. This is probably especially so since the big mergers and the need to do “Metrics” on everything like we made widgits in a factory.
    But speaking from hind sight of someone that started his career before HTS, automated purifications, combichem etc., things were not quite so dull in the lab, and the chemistry was much more diverse. In fact it really stated to get very interesting in the late 70 and early 80’s when a lot of new methods started to be devoloped that made it possible to do all this booring stuff described.

  9. Rick says:

    Following up on Nick K (#6) re: sampling error, it might be slightly more informative to do the same analysis on patents and patent applications.

  10. Anonymous says:

    The publications for a given project within industry do not tell the real story of the entire med chem effort. What you don’t see is all of the crazy ideas that were tried and did not pan out. We all just publish what we select as a nice story out of all the results at the end of the project.
    As for this med chemist, I live for the crazy ideas with the different chemistry. If you work in a good company, I believe this will be supported. The companies that force a chemist to have “clear rationale” for everything made will not support real experimentation. “Clear rationale” is quite the oxymoron in this business. As if we know what will work before we make something, given we have 30 different variables to consider.

  11. Mark says:

    I spent 4 years doing process chemistry and 3 years doing medicinal chemistry for one of the big pharmas. My observations:
    – Process chemistry is pretty fun from a cerebral perspective. While the med chemists make compounds using routes that make analoguing easy, the process chemists get to approach the synthesis from a fresh perspective. Nothing like handing a med chemist a 500 g jar of a bright white solid that was made in 5 steps without a column where before the med chemist’s were screaming that the 0.5 g was THE LAST OF THE MATERIAL. However, you do get opportunities as a med chemist to come up with routes to interesting fragments.
    – However, you do a lot more varied chemistry as a medicinal chemist since so many reactions can’t be run on scale at all. I learned a lot more chemistry as a med chemist than a process chemist. Not surprisingly, process chemistry involves a lot of work that ISN’T bond formation.
    – Having worked in both academic labs and pharma labs, I would say the work is different, but there is NO lack of talent in the pharma labs. I worked with some incredibly bright chemists who would put a lot of the academics to shame.
    – There is something to be said for doing chemistry for practical purposes and not just publication. When I finished my grad work I was proud of banging out a synthesis and proving the value of a methodology, but no one gave a crap I made an alkaloid that had no value beyond being an interesting target. Working out a synthesis and producing material so it could be tested for activity or used in a clinical trial is pretty dam cool.

  12. Anonymous says:

    Don’t have to have complicated structures or chemistry to make good drugs.
    People forget that.
    The only advantage is grabbing new IP space.
    More and more Med Chem is resembling an assembly line.
    My only beef with outsourcing is you place an order, and the person doing the chemistry is not learning anything or contributing to the creative process at all.
    Thus you are entrusting your leaders with a great deal of responsibility.
    Science has always been collaborative, the young bring the new ideas, and the old bring the experience. Challenging old ideas and passing that knowledge was important and fruitful.
    How many times has someone been tasked to making that building block, or that heterocycle, and ended up with 2-3 other side products which at the very least gave some valuable information.
    Or they had a good route and material left over to make other analogues when a bit hit came.
    lets see what WuXi can make for us.
    They have these amines.
    okay good lets just do that.
    But we wanted to test X.
    Can’t do that, not enough people. Look we need to make 50 compounds by July, lets just put the order in. I think this series is dead anyway.

  13. dvizard says:

    Medicinal chemists do medicinal chemistry. Dog bites man.

  14. pharmadude says:

    The target isn’t validated, the assay results might be artifacts, the mouse model might have nothing to do with the human disease, and even if it somehow reaches human trials it could bomb for any number of unknown never to be understood reasons. BUT, you want to spend a quarter of a year doing a 15 step synthesis in the middle of a lead optimization program? How do you justify that time investment? I assume that any given scaffold might have say 4 chemistry FTE years assigned to it to move it through LO. Thats it. If the lead isn’t ready to progress after those 4 FTE years of investment, then its killed. I think most teams have freedom to invest thier FTEs as they see fit, and fast close-in analoging is what they choose as the best back for the buck.

  15. Sundowner says:

    Hmmm, I think some people is missing a point here.
    I am not saying that a) I want my clients to pay me for doing fancy chemistry, b) you have to prepare complicated products with 15 steps or c) that good drugs cannot be prepared in 3 or 4 steps.
    I have done both academical research and industrial research and process development. Industry DOES NOT LACK talent at all. The talent is there.
    But simply, there is a limit about what you can do with the same tools once and again. And as somebody said, if you do always the same, it is not surprising that you obtain always the same results.
    We have developed also products going into clinical with 3 and 4 steps. But that is not the question.
    The question is: are the med chems becoming reluctant to certain synthetic efforts?
    For example, take a look at something like Aliskiren: four chiral centres. Is it too complicated? Does it paid the effort? Some med chems I know would take a look at something like that and say ‘naaa, too complicated, requires fancy chemistry.’
    #14 said: ‘BUT, you want to spend a quarter of a year doing a 15 step synthesis in the middle of a lead optimization program?’ Well, no if I can avoid it. But sometimes you can’t.
    #12 said: ‘The only advantage is grabbing new IP space.’ Amen to that. Problem is, if everybody is doing more of the same, everybody will eventually prepare more or less the same series, which are unfortunately already protected.

  16. AJ says:

    Just like to clarify a couple of points on the manuscript, and I would state up front for clarity and conflicts of interest that I’m one of the authors of the paper under discussion…..
    – To clarify Derek’s point above, neither author is affiliated in any way with the three companies we chose to extract data from. We felt it better to be company-neutral and also to match the prior data on process chemistry, to provide a comparative dataset of med chem vs process.
    – We take (and fully agree with) the comments regarding selection of literature and the inherrent bias. In fact, we do discuss this at length in the manuscript and the limitations this places upon the analysis.
    – Interestingly, the point regarding patents is one we have considered briefly. The publication of the manuscript has invoked some interesting discussions and, from off-the-record discussions with contacts within the larger Pharma organisations, some of these analyses have already been performed, albeit on company-specific datasets. Whilst the actual numbers are understandably different, the underlying trends appear to suggest that there is not a great difference between the data generated in our analysis and othat extracted from the patent literature, contrasting with the views expressed in the comments above.
    – I’m particularly interested in Derek’s comment around “what does it tell us? …..for the most part, we have a pretty accurate mental picture of the sorts of compounds we make. But is that a good picture, or not?” I have my own thoughts on this, and it’ll be interesting to see if they align. But I’ll keep quiet for now.

  17. David Sable says:

    “Outside of changes in reduction state, alcohol-to-halide is the single most favorite functional group transformation, followed by acid to acid chloride, both of which make sense from their reactivity in later steps.”
    Possibly the most enjoyable sentence I have ever read.

  18. MoMo says:

    Reactions are one metric but the unspoken problem in drug discovery is the separation and purification of more polar compounds-those possessing amines or just more polar in nature.
    But that’s why you all make the more hydrophobic drugs-it’s kid stuff.

  19. Mat Todd says:

    Interesting paper, interesting analysis. I particularly like the Bob Silverman quote above (which gets to the heart of it), and the rest of patentgeek’s answer (#7). The reactions used are testament to their robustness, and say nothing about the extensive talent in pharma.

  20. petros says:

    Intersting data for varaiations between companies’ chemistry efforts were presented yesterday by Paul Leeson. Comparing the patent filings from 18 companies over the period 2000-2010 he showed that even when pursuing the same targets there are significant differences with respect to various phys chem parameters, e.g. cLogP and M Wt betweeen companies.
    Incidentally Vertex’ filings had the lowest M Wts of his 18 companies

  21. Indy says:

    What does this study tell us?
    That every body goes after the lowest hanging fruit using the easiest, least complicated routes.
    And once the lowest hanging fruit is gone, now you need to claim a bit higher on the tree to get that fruit. However, this means more work, risk and investment (time and money).
    It is well known that the difficult chemistry is outsourced either to make a more challenging scaffold/chemotype and/or introduce more chemical diversity.
    We have reached a point where convenience is preferred over novelty.
    The reactions mentioned in this study as the preferred ones are the synthetic tools that will allow any chemist under the gun to produce and meet whatever compound quota his/her management order is to be met.
    So we are in a vicious cycle where quotas are to be met and chemists are force to meet them by using the easiest reactions that will enable them get there.
    Want more compounds? Sure boss, no problem!
    Let me nitrate the scaffold, reduce it, and then I’ll acylate it, sulfonate it, do a reductive alkylation, etc.

  22. Hap says:

    Amines, particularly aryl amines, are likely to oxidize and make messes, so even if you can handle them in aqueous solvents, you might not want to.
    Do process people prefer to handle things in aqueous solvents and deal with the waste streams or organic solvents and deal with the solvents (and the smaller aqueous waste streams)?

  23. MIMD says:

    What do medicinal chemists make?
    The future.

  24. Simple_is_beauty says:

    We med chemist set out to look for the simplest answer to a question. If it require complicated strategy, say like protease inhibitors (HIV, Renin, HCV, BACE, you pick it), we do a beautiful synthesis, often better than academic lab. But why will you do things more complicated, if a simple molecule, such as erlotinib would give a good answer? Also look at old drugs, many are simple (

Comments are closed.