[ Skip Navigation ]
Typical Antiviral


As clinical trials of protease inhibitors for fighting AIDS began to show positive results in the early 1990s, process chemists in the pharmaceutical industry faced a formidable challenge. While the studies confirmed that the new drugs blocked the action of HIV in humans, they also revealed that the virus rapidly develops resistance if patients skip doses or don't take enough drug to completely shut down the virus's replication.


SPACE INVADER X-ray structure at 1.9- Å resolution shows a stick model of Crixivan bound to the active site of HIV-1 protease. The enzyme is pictured as a green space-filling model except for the "flap" residues, which are depicted as red and blue ribbons to offer a clear view of the otherwise buried inhibitor.

The required dose of Merck's protease inhibitor Crixivan, for example, was 2.4 g per day--almost a kilogram per patient per year. "We were afraid that if we ever ran out, we would be dooming the patients," says Paul J. Reider. Now vice president of chemistry research and discovery at Amgen, Reider led the team at Merck Research Laboratories that developed the process chemistry route to Crixivan.

"We needed unbelievably high amounts of the most complex product we'd ever made by chemical synthesis," Reider says. "Our success or failure would result in people living or dying."

Earlier, in 1988, Merck scientists had shown that HIV could be rendered noninfectious by replacing an aspartic acid residue in the active site of the protease with an asparagine. Screening of hundreds of molecules to identify inhibitors of the enzyme led to drug candidates that were then improved through computer-assisted modeling studies. Key to the design process were X-ray crystal structures of the lead compounds bound to HIV protease.

"Knowing how the inhibitor binds gives you a tremendous advantage," says Lawrence C. Kuo, senior director of structural biology at Merck. "Designing a drug using X-ray structures is an iterative strategy where you go back and forth with the medicinal chemists. Back then we could only do five to eight crystal structures a year, which fortunately served the purpose."

A similar kind of back-and-forthing was going on among the medicinal chemists and process chemists, according to R. P. (Skip) Volante, Merck's vice president of process research. Volante was the project leader for Crixivan process research during the drug's development.

"We needed to get a process in place quickly to speed the drug's registration," Volante says. "As a treatment for a life-threatening disease, it would move rapidly through fast-track approval at the Food & Drug Administration. We needed an efficient process capable of making metric tons of material. It had to be scalable and cost-effective. The complexity of the molecule, with its five asymmetric centers, made it a particular challenge. Out of 32 possible stereoisomers, we had to make a single enantiomer in pure form."

Crixivan wasn't Merck's first protease inhibitor candidate, Volante notes. In working on the initial compound, Volante's team devised a synthetic strategy for putting a basic framework in place. The strategy was fed back to the medicinal chemists, who used it to make variations on the core structure.

"Early on, we devised chemistry for making cis-aminoindanol, which was used in the first compound and turned out to be an important piece of Crixivan too," Volante says. "Later, we found we could also use cis-aminoindanol as a powerful chiral auxiliary to direct the stereochemistry of two subsequent reactions. Usually, chiral auxiliary agents are wasted, but ours was built right into the drug molecule."

With continued improvements, the process chemistry team cut the time to synthesize each batch of Crixivan to six weeks and improved the overall yield from about 15% to 50%. By 1995, however, as construction of two plant facilities got under way, Merck's pilot plant was barely keeping up with the production needed for clinical trials. The company came under fire from AIDS activists who accused the firm of shying away from supplying the compound to patients on a "compassionate-use" basis. Reider agreed to allow an outside expert hired by the National AIDS Treatment Advocacy Project to analyze the firm's efforts.

"My duty was to ask very awkward questions," says the consultant, Trevor Laird, managing director of Scientific Update and editor of Organic Process Research & Development. "I was given complete access to everything Merck was doing. I came away feeling that the company was on an aggressive timeline and it would have been extremely difficult to go any faster."

Despite the uncertainty of supply, Reider agreed to make Crixivan available for compassionate use. A year later, Crixivan was approved by FDA, not long after the first protease inhibitor, Hoffmann-La Roche's Invirase, and the second, Abbott Laboratory's Norvir. Merck's plants soon came on-line, and the supply of Crixivan has never been interrupted.

Today, protease inhibitors are generally used as part of a combination of three or more drugs. A newer generation, including Abbott's Kaletra, promises significant improvements in efficacy and safety.

The effort to bring Crixivan to market was huge. "Everyone on the team, even Merck's vendors and suppliers, worked with such clarity of purpose," Reider remembers. "It wasn't about money; it wasn't about credit; it was just about saving people's lives."—PAMELA ZURER


The Top Pharmaceuticals
That Changed The World
Vol. 83, Issue 25 (6/20/05)
Table Of Contents

Indinavir Sulfate

Crixivan structure


  • 2,3,5-Trideoxy-N-[(1S,2R)-2,3-
    -D-erythro-pentonamide sulfate

CAS registry

  • 157810-81-6

Other Name

  • Crixivan

Did you know that FDA approved Crixivan just 42 days after receiving Merck's application?


  • $294 million in 2002