Ask star antibody developer and Medicxi partner Kevin Johnson to name the biggest challenge in bringing a new antibody into the clinic, and his immediate answer has only three letters: CMC. Everyone knows how big a challenge it is to get the manufacturing piece right for an antibody. And many promising projects stumble due to insufficient focus on the deceptively simple process of actually making the drug.
The difficulties are obvious to most drug developers – almost all of whom have a biology background. Proteins are orders of magnitude more complex than a typical small molecule chemical drug. They can vary in structure (for example due to glycosylation), be unstable (thanks to oxidation of methionine or deamidation of glutamine), change conformation and so lose activity, aggregate and fall out of solution – in short, the challenge of making reproducible lots of any protein is transparently substantial.
Contrast that, though, the with attention typically paid small molecule manufacturing. “Chemistry is just cooking” one biologist and biotech CSO once told DrugBaron dismissively – only to discover a year and £2million later that his company’s deceptively simple lead compound was every bit as challenging to make at scale as any antibody.
“Making a small molecule drug at scale is indeed ‘just cooking’” according to David Fox, who heads up chemistry for RxCelerate, ‘but like cooking, most of the skill is in finding the right recipe in the first place, rather than just following it. Fundamentally, finding a suitable route to manufacture any novel molecule on a kilo, or larger, scale is a chemistry research activity – and in research outcomes are never guaranteed.”
That’s quite different from the usual perception. To a biology-trained drug developer, if you can make a gram of your molecule, the task of making kilos or tons seems like an engineering problem rather than a chemistry one. And that misperception often leads to trouble, when they engage a contract manufacturer staffed by chemical engineers, who know very well how to drive the plant, but fail to properly understand the details of the underlying chemistry. “Chemical engineers are engineers who work with chemicals – not chemists” says Fox.
“Finding a good, scalable manufacturing route for a new compound has more in common with medicinal chemistry than with engineering. In a medicinal chemistry project, we are looking to choose one molecule out of an almost infinite set of possibilities that has a particular set of properties. In a scale-up project, we are looking to choose one route, one process out of a similarly large number of plausible variations.”
No-one starts a medicinal chemistry project with the expectation that success is guaranteed. It may very well prove impossible to have all the desirable features in a molecule, and most often the question becomes ‘can we find one that’s good enough?’. Nor would many people attempt such a difficult task without a team of experienced medicinal chemists (whether employed in-house or out-sourced). Yet those same people take a completely different approach to scale-up, viewing success as simply a matter of spending the time and money, placing the contract with the cheapest bidder many of whom lack the necessary skills for what is, in essence, a research project.
The situation is made worse by contract manufacturers who exploit this common misperception in their client base. No matter what molecule you draw on a piece of paper, or how little is known about possible synthetic routes, you can get a quote for a contract manufacturer to make you a kilo of it. That re-enforces the view that success must be inevitable. After all, for a project manager with no chemistry experience, holding a piece of paper with a price and date for delivery in their hand, it looks no harder than going to a supermarket to buy a loaf of bread.
No-one sensible would write a quotation to discover a novel development candidate molecule from scratch for a fixed fee with a delivery date. So why expect that for finding a scalable manufacturing process where one does not already exist? The answer is simple: because clients are willing to believe it is straightforward, then companies secure their business by confidently telling a fairy-story that represents what is actually a journey of scientific discovery as little more than shelling peas.
“The consequences are predictable” opines Nigel Ramsden, who has over twenty years experience in the chemistry of drug development, at companies large and small, and who currently co-leads the chemistry division at RxCelerate with David Fox. “More times than not, the first attempts fail, timelines extend as ‘deadlines’ are missed, and the budget expands uncontrollably as things get repeated to find a suitable process.” Having started down the pike the client soon finds themselves tied in with little or no leverage to control costs or dictate time lines.
Worse still, the people trying to solve the problems often don’t have the necessary skills. The constraints imposed by the engineering capabilities of the physical plant are well understood but the constraints imposed by the chemistry itself are relegated to a distant second, if they are considered at all.
“We have seen numerous examples” says Ramsden. “Even when you give the contract manufacturer a working process, things go wrong because of their poor understanding of the chemistry. Even small things can make a big difference: one company started a batch process several hours late, and so introduced an overnight hold into the process, and were surprised to discover new contaminants in the product. But even a modest knowledge of chemistry predicted exactly what would happen.”
The lesson is simple. Unfortunately it took DrugBaron almost a decade to learn: scale-up for small molecules is, more often than not, a research activity. Questions such as how much will it cost or how long will it take are less important than who are the right people to help me find the right process. As with any research, there can never be a guarantee of success, but the chances of emerging with a good process, in tact timelines and a balanced budget are so much higher with chemists guiding the process engineers rather than being seduced by a fixed price, and usually altogether fictional, quote to deliver kilos of your compound from a contract manufacturer.
Kevin Johnson’s mantra for biologics, it turns out, is just as applicable to small molecules.
RxCelerate Ltd is an outsourced drug development platform based near Cambridge, UK. We specialize in delivering an entire road map of drug development services from discovery and medicinal chemistry through to formal preclinical development and clinical up to Phase IIa. In the last five years, we have witnessed dramatic changes in the drug development …