Decades after the petrochemical and chemical industries began, and still continue, to reap the multiple benefits of continuous processing (CP), at last the pharmaceutical industry is now looking to do the same. It can hardly afford not to. At a time when the industry is coming under ever increasing pressure from low cost competitors in the East, as well as from burgeoning environmental and other legislation back at home, running chemical reactions in miniature, high throughput, micro reactors and other continuous flow systems promise not only to speed up production, reduce inventory and operating inputs, but also to bring down the costs of manufacture, CP advocates contend.
On this last point, however, the pharma industry is staying remarkably silent. Anything that has to do with bringing down manufacturing costs could lead to the possibility of lower drug prices, which is a discussion the pharma industry definitely does not want to have, one industry insider who wished to remain anonymous told C&I. While all of the major pharma companies are known to be actively investigating the possibilities for introducing CP into their various manufacturing processes, few are prepared to discuss openly exactly where and when they are likely to implement the technology.
At Pfizer, the company has an active programme developing continuous and semi-continuous processes for intermediates and active pharmaceutical ingredient (API) as well as for drug product, says Dan Pilipauskas, director of the API development team.
A drug API made by fully continuous manufacturing at Pfizer, however, is more than two years away, he estimates. Rumours that GSK was about to launch a drug API made by CP, meanwhile, surfaced earlier this year at an SCI conference on the subject (C&I 2007, 6 , 5), but the company swiftly denied the claims, and did not wish to be interviewed for this article. Another contractor for the agrochemicals sector also withdrew from the article just ahead of publication to protect client anonymity. It is possible, however, Pilipauskas speculates, that a drug made by CP is already available on the marketplace, but not widely publicised.
For companies looking to embark on CP manufacturing, the potential cost savings could be significant. The initial capital cost of starting up a new CP plant can be 25-40% cheaper than installing a traditional batch plant, for all capacities even down to low capacity plants making less than 10t product/year, engineering contractor Foster Wheeler estimates. Even greater cost savings are likely to accrue during the lifetime of the plant, meanwhile, thanks to lower energy bills, reduced staffing and maintenance and better product consistency, not to mention an up to 40-fold reduction in utility requirements for industrial gases, purified water and other inputs.
Ultimately, the decision whether to run a CP operation comes down to a simple cost-benefit-risk assessment, says Foster Wheelers senior principal pharmaceutical engineer Huw Thomas. If your current batch process is economic and efficient, and if the only benefit is capital expenditure of new plant and a new plant is not needed then why build one?
While most of the continuous pharma and fine chemical processes currently in operation do not use CP for all processes stages, even for hybrid batch-continuous processes there can be substantial benefits. It is estimated that 50% of batch processes are currently heat or mass transfer limited, and a hybrid approach can be used to overcome these limitations and build up continuous experience without the investment/risk of moving to a fully continuous process.
Hundreds of papers have been published showing reactions can be more efficient (yield, selectivity) where the inherently better capabilities (mass and heat transfer, plug flow profile) of continuous reactors can be utilised to make the chemistry more efficient. Likewise other continuous unit operations such as counter-current extraction and thin cake filtration/drying give inherent benefits in terms of capacity and efficiency, says Thomas.
But while big pharma companies are reluctant to give away too many details, many of their fine chemicals contractors are only too keen to promote the CP work they are doing. At UK-based fine chemicals firm Phoenix Chemicals, for example, CP has been at the heart of the companys operations ever since it was founded in 1991, says technical director Lee Proctor, and today there is there is a CP element in 100% of our products; all of the products we make involve one or more CP steps.
The companys mission is to provide engineered solutions to access difficult to handle chemistries, he elaborates. Using CP offers an obvious advantage, because the safest way of handling a hazardous material is by using as little of it as possible and generating and using it fast.
One example is a CP route involving the potentially highly toxic and explosive diazomethane gas to make various intermediates for HIV-protease inhibitors, one of the companys core product lines. The plant carries out four reaction stages using a fully integrated and highly automated telescoped CP reaction platform, Proctor explains, incorporating several analytical sensors that provide continual feedback on the mechanical and chemical operation of the plant. The facility allows us to operate the plant at a true steady state condition and to finely tune the process to minimise unwanted side reactions which ensures robust high quality product consistency.
All of the CP reactors that Phoenix uses are designed in house, and include a novel multi-purpose plug-flow reactor system that allows researchers to vary the residence time of the reactants in the reactor, by stretching or compressing at will the length of the reactor tube. Along with the amount of mixing and heating and cooling of the reactor, the residence time of materials spent in the tube is a critical factor determining product quality and consistency, Proctor notes.
Avecia spin-out Reaxa is another UK firm looking to capitalise on the growing interest in CP. Formed in 2005, the company specialises in producing a range of encapsulated catalysts and supported metal scavengers for use by pharma customers. Both product lines are resin-based and ideal for CP use, says Michael Pitts: From the start we wanted to exploit the technology in flow applications.
Performing standard catalytic reactions in flow systems is notoriously difficult, he explains, because once the catalyst has performed its function, it will usually spit out its active metal, which can crash out of solution, clogging up the reaction tubes. In our case, the catalyst is bound within the resin, he continues, effectively trapping the catalyst within the system and preventing any leaching of metal.
Reaxas QuadraPure resins work as metal scavenging agents to mop up the spent metals generated by catalysis. The usual way people use them is by dumping the catalyst resin in the reaction mixture and filtering the loose resin off after the reaction is complete, Pitts says. For use in CP systems, however, Reaxa has recently launched the catalysts as a more convenient pre-packed cartridge format for use from lab to pilot plant scale.
In early lab scale tests using the cartridge, we have seen significantly increased kinetics with the catalysts and increased rate of scavenging and achieved higher loadings 25% more metal at least is extracted using the flow cartridge with the QuadraPures. The ease of use and handling, not putting loose solids into reactors, are big advantages.
Running processes in flow means a smaller reaction area, which has safety benefits but also allows use of novel technologies, such as microwave heating, that are difficult to apply to a large tank.
Pilot scale evaluation of the QuadraPure cartridge is expected to get under way later this year, he adds. Several of our customers are evaluating the encapsulated catalyst technology in CP applications.
UK speciality chemicals company Thomas Swan is also hoping to make good its CP expertise over in the pharmaceuticals sector. A pioneer in the field of supercritical fluids (SCF), it operates the worlds first multi-purpose continuous flow SCF plant, with a capacity of up to 1000 t/year.
SCFs are inherently geared towards continuous flow systems because of their novel flow properties, says business manager Russell Clarke. SCF technology offers chemists a powerful tool for the simplification of chemical processes as well as facilitating improvements in yield and selectivity over batch processing.
The firm has already discussed the use of SCF processing with several pharma firms, he says. Hydrogenation is widely used in the synthesis of pharmaceutical intermediates and it is hydrogenation where Swans initial activities have been focused as it has demonstrated the most promising results.
The overall potential for CP technology in pharmaceuticals is considerable, says Derek Lindsay, director of manufacturing innovation consortium Britest. Our data for the fine chemical/pharma sector suggest that about 5-10% of reactions should only be done continuously, 40-45% could be done continuously, if economically beneficial, and 40-50% should still be done batch.
Whole process design
A not-for-profit organisation run by its members including most of the major pharma and fine chemical companies Britest was established in 1998 with the aim of helping companies to develop the best processes and manufacturing strategies for their businesses. It is currently working with most of the major pharma and fine chemical companies on CP processes, which can increase productivity by anywhere from 5 to 300%, Lindsay notes. However, the emphasis is on whole process design, including work up, isolation and formulation, rather than on individual reaction steps, he stresses. The Britest approach is to understand the fundamental drivers of the process, then decide on the best manufacturing technology, which could be batch or continuous. They are not mutually exclusive.
Whether CP can live up to all its current promises, however, is a question for debate. At AstraZeneca, director of process chemistry David Lathbury says the company does not currently have any manufacturing processes using CP as far as I am aware. Researchers recently evaluated two possible CP reaction routes, he adds, but in the end decided to stay with batch as this was cheaper. The volumes being made were relatively small so it didnt make financial sense.
Contrary to other commentators, Lathbury believes that CP is unlikely to have much impact in bringing drugs to market more quickly. The idea that researchers can simply scale up standard laboratory syntheses to full scale production by increasing the number of reactor modules is one that has tantalised chemists since the prospect of CP was first introduced several years ago. In reality, he argues, it will have little effect since often the most time-consuming stage is the reaction optimisation stage for which there is no way of getting around.
In some ways CP makes the early drug development phase harder as it presents us with another choice: Twice as many choices but only same amount of same amount of time to work through them all.
Re-registration of drug processes is another obstacle to developing new CP routes for existing batch processes. Worldwide registration can cost up to $20m and take up to two years, Lathbury says. CP is often viewed as a battle between engineers and chemists. He continues. Theres a temptation for engineers to look at chemists as luddites who wont embrace new technology but from our side we look at engineers as a one trick pony only interested in pushing CP for all reaction types.
Where CP will become important, however, is when the economics start to change, Lathbury concedes. If drug prices decline, as expected owing to political pressures etc, then the economics start to change and you have a situation where you start to look at CP for second generation, higher volume, processes for which CP becomes more attractive.
This article is partly based on an SCI meeting on CP in the pharmaceutical industry in March 2007