Although he's a doctor of medicine, not chemistry, Peter G. Milner understands the importance of chemistry to drug discovery. After all, his company is built on it.
Milner is president, chief executive officer, and cofounder of ARYx Therapeutics, a Santa Clara, Calif., start-up focused on developing small-molecule drugs that overcome the design and metabolism problems of existing therapies. Its lead product, ATI-2042, is designed to work as well as the $750 million-a-year arrhythmia treatment amiodarone, but without the adverse side effects caused by accumulation of amiodarone in the body.
Born in England and educated there and in the U.S., Milner is steeped in the pharmaceutical industry. His family is the majority owner of ML Laboratories, a publicly traded company based in the north of England, and Milner cofounded the cardiovascular drug company, CV Therapeutics, in 1992, after four years as an assistant professor of medicine at the University of Washington, St. Louis.
Milner understands the appeal of the biotech side of the drug industry--early in his career he discovered, cloned, and sequenced an antigenic growth factor. However, he says his appreciation for chemistry has grown over the years. "When I left CV in the 1990s," he says, "I was looking for something that made medicinal chemistry simple, explicable, and predictable. Everybody was going off into high-throughput screening and automated discovery. I thought that was a mistake."
While scouting for venues to put his thinking into practice, he met a French medicinal chemist, Pascal Druzgala, who had nurtured a concept the two now call retrometabolics--the design of molecules prospectively so they metabolize into inactive, nontoxic, water-soluble by-products.
Milner and Druzgala founded ARYx in 1997 to exploit retrometabolics in the development of safer drugs. For its first few years, the company was funded privately by Milner's family and friends. In 2002, it attracted $25 million in private equity from a group of investors led by MPM Capital; Milner is now in the process of raising another $50 million.
Not surprisingly, drug chemistry is important to Druzgala and the seven chemists working under him. "Conventional chemists design drugs to make them active and then worry about what the body does with them after they've worked," Milner says. "We think about metabolism at the same time as we think about the drug."
Druzgala's team will start with a drug that has been removed from the market for safety reasons or that is on the market but has side effects. They redesign the drug by first determining ideal metabolites and then adding chemical diversity back to the metabolite to create anywhere from 15 to 100 analogs. From these, they select a drug that is as potent as the original but has a safer metabolism.
Milner says ARYx chemists integrate process chemistry into this design process with an eye to creating a commercially viable synthetic route. But ARYx isn't a chemical manufacturer, so two years ago, in anticipation of needing current Good Manufacturing Practices (cGMP)-qualified quantities of ATI-2042--and needing it fast--Milner looked to outsource.
He knew of the Manchester, England-based process development firm Ultrafine because it is based near his family's company and because Michael Clark, an investor in Ultrafine, is a friend of his. Druzgala contacted Ultrafine and was impressed both by its responsiveness and by the way Michael Harris, Ultrafine's head of business development and a Ph.D. chemist, grasped ARYx's chemistry needs.
ARYx hired Ultrafine in July 2002 and just three months later took delivery of the first kilogram of ATI-2042 for preclinical work. After another five months, Ultrafine delivered multiple kilograms of the drug made to cGMP standards. Harris says this kind of rapid turnaround isn't unusual. "ARYx is a typical customer in that speed is of the essence for them," he says.
SPEED COMES at the expense of major synthesis redesign, Harris acknowledges, and Ultrafine met ARYx's timeline by sticking fairly closely to Druzgala's original medicinal chemistry route. Ultrafine chemists, aware that their process wasn't economical for ton-scale production, did rough out what they thought would be a viable commercial synthesis.
But with its focus on lead optimization up to cGMP production for Phase I and II clinical trials, Ultrafine isn't equipped to handle commercial production of a potentially high-volume drug like ATI-2042. This is okay with Milner. "Some companies run their small-scale businesses as a loss leader," he says. "Ultrafine does it for a living and does a pretty good job at it."
Indeed, ARYx has since contracted with a larger, U.S.-based company that has built on Ultrafine's work to design a new, commercial-scale ATI-2042 synthesis. Production costs that were $100,000 per kg in ARYx's labs are now down to $10,000, and Milner predicts they will be less than $1,000 per kg at the ton scale. He notes that a robust synthesis is critical in negotiating a licensing agreement with a large pharmaceutical company. "It can be the difference between a $15 million deal and a $40 million deal," he says.
The niche between drug company medicinal chemistry and commercial-scale pharmaceutical chemical production is often overlooked, but it has been a successful one for Ultrafine. Harris says the company has expanded rapidly since receiving a venture-capital injection in 1998, with sales growth last year alone of 50%.
Loyal customers are part of that growth. ARYx's second drug in development is ATI-7505, a safe analog of the heartburn medicine cisapride. Johnson & Johnson sold cisapride for several years as Propulsid but, in 2000, voluntarily removed it from the market because of interaction problems.
Soon after ARYx placed ATI-2042 with Ultrafine, it outsourced ATI-7505 to a small company in California. That firm "did a terrible job," Milner says, and is out of business today. ATI-7505 has now been scaled up by Ultrafine.
GROWING UP In the past two years, Ultrafine has opened a 25-hood addition and a 200-L kilo lab.