A means to an end

When a chemist is hired fresh out of a university into the pharmaceutical industry, a number of mental adjustments have to be made. Some of them are pleasant, such as the sensation of actually being paid in real money. The facilities and resources at one’s disposal are also usually better than most academic groups can claim, and getting used to that is no great hardship, either. For example, I remember being struck by the ’don’t make what you can buy’ rule when I started out in drug discovery, because until then I’d worked under that law’s exact opposite. But some of the other changes are more subtle - and for some researchers, they can be more troubling.

The chemistry itself can be one thing that takes some getting used to. For those scientists who have done high-level total synthesis projects, there can be a certain letdown after seeing the kinds of reactions that are common in most drug labs. Say farewell to the asymmetric retro-Zambezi rearrangement at step number 28, and say hello to plenty of common-or-garden palladium couplings, amide formations, and reductive aminations. We like those reactions in the industry because they make useful structures, and because they’re easy to run and tend to give some level of product. 

Tricky, difficult reactions that require excellent hands and exacting conditions make for a fine PhD thesis, but they’re the opposite of what a drug lab needs. For some people, the thought of never again being on the tight-rope at step 28 will come as a relief (it did to me), but others may miss the excitement. It’s safe to say, though, that starting a synthetic scheme of that complexity in a pharmaceutical lab is at best ill-advised, and, at worst, grounds for dismissal. The shorter and less eventful the synthesis of a drug candidate, the happier everyone is. 

Some chemists will need to modify their attitudes towards the yields of their reactions as well. In a total synthesis, a 50 per cent yield at a late stage can be a disaster worth spending months to try to avoid. In the early stages of drug discovery, though, while a high yield is always desirable, there are really only two yields: enough, and not enough. Enough can easily be twenty milligrams, and if that somehow came from a gram of starting material and represents a 2 per cent yield, well, that’s unfortunate, but you’re still going to send that compound in for testing. You’re going to send everything that looks halfway reasonable in for testing, no matter where it came from. If something interesting turns up, the time will come to worry about how to make it more efficiently.

Say hello to common-or-garden reactions 

The sorts of compounds being made are different as well. New researchers will be saying hello to nitrogen, too, and particularly to basic amines and nitrogen-containing heterocycles. The huge majority of drug candidates are nitrogenous, some of them a bit too nitrogenous for their own good, frankly, and working on them can be a big switch for chemists who’ve been doing, say, marine natural products or carbocyclic rearrangements. Heterocyclic chemistry doesn’t get taught much as a separate subject any more, and often doesn’t make it into the highest-end journals, since it’s considered to be such a well-worked field. Many are the times, though, when a drug discovery programme ends up depending on the sorts of chemistry that you’d have flipped right past if you’d come across it in a journal six months before.  

These differences all have to do with the purpose of the chemistry. In many academic projects, chemistry is an end in itself. An unusual or beautiful process is worth doing because it’s unusual or beautiful. Drug research, though, is applied science, and things are generally only considered worth doing if they produce something useful. Useful, in this context, is easily defined as ’producing a compound that doctors, patients and health plans will pay money for’. Anything that leads to this state of affairs more quickly and easily is probably good, anything that leads away from it or makes it more difficult is very likely bad. Some new researchers find this sort of clarity welcome, while others find it a bit more than they were looking for. 

The above definition leads to the realisation that chemistry, in a pharmaceutical setting, is a means to an end, and (depending on the project) not necessarily the most important one. Without the medicinal chemists, it’s true that everyone downstream would have very little to work on - but without the other groups, the chemists would never know if they’d made anything worth making. We all depend on each other. In the end, the adjustments that new chemists have to make in the industry all come from this one: that they’re part of something larger and even more challenging than what used to be their whole scientific world. 

Derek Lowe is a medicinal chemist working on preclinical drug discovery in the US