Process development and the need for innovation

Essential but unfashionable, process development lags way behind other areas of the pharma industry in the innovation stakes, as Peter Nightingale explains to Sarah Houlton

Process development is the unfashionable part of pharmaceutical manufacturing. Successful discovery of blockbusting new chemical entities gets plaudits from shareholders and analysts, while the general public sees prettily-designed packaging and multimillion pound marketing campaigns. Even the skills of product formulation are widely acknowledged, with line extensions adding to potential revenues and simpler-to-take products making administration to the patient easier.

But without the skills of the process development chemist, there would be no products. Reactions that work beautifully in the laboratory rarely work adequately on an industrial scale. Reagents may be too expensive, the chemistry could be too dangerous, or the reaction just may not work in a 1,000l-plus reactor. In the past, manufacturing costs have represented only a small part of the price of a pharmaceutical product but, as drugs become more complex chemically, and managed health care organisations are putting pressure on margins, their significance is increasing. Yet sufficient impetus for creating innovative chemical manufacturing technology is not there, and potential long-term cost savings are being missed.

As Dr Peter Nightingale of the development company Synprotec explains, process development can either be considered as the last part of research, or as the first stage of manufacture. 'Without exception, every pharmaceutical company I talk to, including major multinationals, agrees that process development is the first stage of manufacturing, not the end of research. Yet in most cases the structure and organisation of the development function is more closely related to research.'

Nightingale believes the underlying cause of the problems facing process development is that there has been insufficient real innovation in the sector for decades. 'Process development should not be confused with chemistry, where there have been huge innovations, with new reagents, new reactions, advances in organometallic chemistry, chiral catalysis and so on. But the way they've been used hasn't really changed.

'It is very difficult to imagine any other high-tech sector where the point-of-production equipment has changed so little. Look at a car factory, a food plant, or even a farm; production used to be centred on the skills of its operators, whereas now it is heavily dependent on modern equipment. But if you were to put a chemist or an engineer from a 1950s factory into one of today's chemical plants, as long as he was reasonably competent, within a matter of days he would be capable of running it. Granted, there has been some modernisation such as computer control systems and data logging, but the production is still largely centred on the basic stirred, jacketed batch reactor.'

This contrasts strongly with research, where the technology changes so rapidly that, within a decade or so, techniques are superseded. 'The key to process development is the synergy between the equipment and the chemistry,' he says. 'Just because the chemistry itself has continued to be innovative, it does not mean there have been parallel advances in process design. The target of the process development chemist has traditionally been to fit novel chemistry into old-fashioned plant.

'One of the reasons that the plant hasn't changed in 50 years is that it's been successful. It's worked. And flexible general-purpose plants have an obvious attraction in a sector where product lifetimes are difficult to predict to say the least. However, process development chemists and process engineers have to face up to the fact that, just because what they do has been successful for the past 50 years, it doesn't mean it will continue to succeed indefinitely.

But what can be done to improve innovation in process development? 'There is a fundamental requirement for good process development chemists and process engineers who talk the same language,' Nightingale explains. 'This is not as easy as it sounds — their careers and their education tend to push them apart, and chemical engineers and chemists historically have not co-operated well. Chemical engineers need to become more involved in processes at the early stages of development.

'With manufacturing technology, one of the key requirements is real-time analysis. There is no point in having a reaction that's complete in seconds it if takes half an hour to take a sample and run it on the GC. There has to be analytical feedback in real time.

'Faster reactions save time and money, but, because the equipment is not designed to cope with the heat transfer requirements of a very fast process, the skill of the development chemist is to slow it down to fit the available plant. Yet the reaction may be able to go to completion in seconds, although it would destroy the entire factory in the process. Hence slowing the reaction down to take, say, a standard 8hr. That's not good enough. If a reaction can be completed in a few seconds, surely the process development challenge should not be how to fit it into the plant, but to create plant that can cope with very fast, highly exothermic reactions.'

Finding a solution to this dilemma will require different designs of reactor, allied to real time analysis. A number of groups, including collaborative ventures, are exploring novel reactor designs, from relatively simple concepts such as loop reactors, to more radical approaches such as microchip reactors and other applications of nanotechnology. 'The ideas are there,' says Nightingale, 'but they've not really been taken up or even taken seriously enough. How can we break the mould of how things have always been done? I think it's important that the process development chemists and process engineers start looking for solutions — we've got to think of other ways of manufacturing chemicals rather than just using conventional multipurpose plants.'

Short-term financial pressures are an important factor. It is easier to fit new products into old plant. The pharmaceutical sector presents a paradox, because despite the overall long-term approach the sooner the product can be put on the market, the sooner it makes money and the longer the company has to exploit it before its patent runs out. This works directly against the need to develop newer, more efficient processes, as companies are reluctant to spend big money on developing a product that may not reach the market, so they go for the quick option of adapting the process to fit existing plant designs. The situation can be even worse after product launch since changing the process also costs money, as it needs to be re-registered.

Nightingale believes that the solution to this paradox is the key to success of the fine chemical industries in the coming century. 'Unfortunately, I don't have the complete answer,' he says. 'First of all, we need to get people to realise there is a serious problem. There are no quick answers, and the advent of modern discovery techniques is bringing so many lead compounds forward that the time pressure to bring products to market leaves ever less time for innovative process development.

'In the short term, this might represent good news for specialist outsourcing development companies like Synprotec. But in the long term the risk must be that the manufacturing operations of the life science industries will end up in a similar situation to the cotton industry in Manchester. I'm sure 150 years ago, the owners of the cotton mills were convinced no one could ever make cotton as well as they did. But they didn't modernise their production methods, and saw their industry migrate to cheaper economies abroad. 'If the Europeans and the Americans are going to compete with Indian and Chinese companies, where there is no shortage of good chemists, there's going to have to be a change in manufacturing technology because, on pure cost, they are never going to be cheaper.

This doesn't necessarily just apply to pharmaceuticals and biochemicals, it applies to all manufacturing industry. Some people take the view that manufacturing is unimportant and will all move eastwards anyway, and we are going to live in some sort of non-productive knowledge based economy. I don't believe that works; I think a successful economy has to have a manufacturing base. When all the manufacturing is gone, the knowledge and innovation will follow and we will be left with nothing. An economy needs to have a manufacturing heart.'