Combinatorial Chemistry Takes On Process Development

Combinatorial chemistry — the simultaneous creation of tens to thousands of new chemical entities — promises to speed drug development by multiplying pharmaceutical chemists' hands by factors of ten. We have not yet seen the impact of combinatorial technology in the marketplace or even for pipeline drugs, probably because combinatorial strategies are only a few years old. And given the time and financial resources required to introduce new drugs, the hope of more, better drugs may fall beyond the capabilities of even the largest pharmaceutical companies.

Combinatorial chemistry may, in fact, have a much more profound — and near-term — impact on process development. While development scientists have no desire to synthesize 100 × 100 arrays of unique, new peptides or oligonucleotides, they could certainly use the multiplicity functions of 20- to 100-reaction combinatorial systems to optimize such factors as temperature, solvent, addition/mixing order, and other nitty-gritty components of development and scaleup.

According to Michael Flavin, president of MediChem Research (Lemont, IL), a contract research/manufacturing company, a connection between automated parallel synthesis and pharmaceutical chemical development "definitely" exists.

MediChem uses a Bohdan parallel synthesizer for process and scaleup research.

"In drug discovery the objective is to make as many different compounds as quickly as possible," Flavin said. "But eventually a company settles on a lead compound which it wishes to take to preclinical and clinical trials. To get to that point companies must sort through a lot of variables. But now that they're focused on one compound the focus shifts from how many different compounds can be made to how well the one or two lead compounds can be made. At this point process R&D takes over to find the best synthetic conditions. So here, companies are screening not for molecular diversity, but for 'reaction space' — all the parameters that can possibly be used to optimize processes for making the target compound.

"Using automated synthesis, companies can uncover a vastly larger array of reaction conditions within a shorter period of time [than they could by manual synthesis], ultimately arriving at optimal reaction conditions giving the highest yields and purities."

Instrument companies agree that parallel synthesis is not just for drug discovery.

"Combinatorial chemistry systems could certainly be used in manufacturing mode," says Argonaut Technologies Marketing Director Ed Long. Argonaut specializes in products for accelerating the discovery of new organic molecules, particularly pharmaceuticals. The firm's products include proprietary synthesizers, database software, solid-phase supports, and consumables. Argonaut's flagship synthesizer, the fully automated Nautilus 2400, supports reactions ranging from 5 ml up to 23 ml in volume. Vessels have a side port which enables sampling in mid-reaction. Argonaut also offers a manual synthesizer, the Quest 210, which runs 20 reactions at once.

"Although our products are primarily targeted for chemistry [new compound] development, several customers have begun using Argonaut synthesizers for process optimization and characterization," Long said. "The idea is to take the lead compound and use the instrument to study factors that affect yield, such as temperature and pH."

Argonaut's Quest 210 is an all-purpose parallel synthesizer.

Most combinatorial/automated synthesizers use solid-phase chemistry, a technique first used to automate peptide synthesis. Solid phase reactions are ideal for unattended synthesis of milligrams of a target molecule but unsuitable for process development and optimization. Argonaut's synthesizers perform admirably either with solid- or solution-phase reactions.

Zymark Corp., another firm in the combinatorial chemistry business, approaches this emerging market not from chemistry but from robotics and automation. In this regard Zymark's instruments appear more naturally suited to process development than, say, Argonaut's (whose focus, after all, is chemistry). Exploiting its existing technology base for positioners, robotic arms, dispensers, and other staples of laboratory automation, Zymark has assembled a Solution Phase Chemistry System that combines weighing, reagent addition, reacting, capping and uncapping, filtering, liquid-liquid extraction — all the operations normally performed by hand.

Strong on robotics, Zymark's parallel synthesis strategy carries out discrete manipulations in addition to running reactions.

The Solution Phase system performs reactions in screw cap sealed tubes which can be taken from reflux to -30º C. The extraction feature is especially useful to process development chemists seeking to optimize that critical aspect of product isolation.

Zymark produces reactors capable of creating unique reaction conditions for each tube, but that's not always needed. Brian Lightbody, vice president of drug discovery business development at Zymark, suggests that reaction optimization may not be the only reason process chemists would use parallel synthesis. "Optimization aside, parallel synthesis offers development chemists an easy way to obtain statistically valid data for a single reaction. Once a reaction is optimized," Lightbody points out, "it needs to be validated by collecting multiple data points," a task for which parallel synthesizers are eminently suited.

At what point will parallel synthesis be accepted by most pharmaceutical process engineers? MediChem's Flavin believes that will only happen when synthesizers and analytical instruments are connected, fully automated, and computer controlled. "One could envision connecting these components together right now," Flavin said, "but connectivity must be within a statistically valid framework and under computer control."

By Angelo DePalma

For more information: Michael Flavin, President, MediChem Research, Inc., 12305 South New Ave., Lemont, IL 60439. Tel: 630-257-1500.

Ed Long, Argonaut Technologies, 887 Industrial Rd.., Suite G, San Carlos, CA 94070. Tel: 650-598-1350, ext. 234.

Brian Lightbody, Zymark Corp., Zymark Center, Hopkinton, MA 01748. Tel: 508-497-2232.