Pfizer's Ann Arbor laboratory is the largest private-sector employer in the Michigan college town. It is also the birthplace of Lipitor, a cholesterol treatment whose annual sales of $13 billion make it the biggest-selling drug on the planet. So the lab's 2,100 staff were stunned when Pfizer announced on 22 January that it will close the lab by the end of next year.

The closure is part of a move by the world's largest drug company to cut between $1.5 billion and $2 billion from its annual costs, eliminating some 10,000 positions, or about 10% of its workforce. That will include trimming its research and development staff from around 14,000 to 12,500, according to Pfizer's global research chief John LaMattina, and consolidating researchers at four key sites.

The changes will make Pfizer's research operation more flexible and cost-effective, says LaMattina. “The simplification will add a lot to our efficiencies as well as the speed of our decision-making.

It takes decades to build something like that but, evidently, only 18 months to dismantle it.

But Michigan academics were left reeling by the announcement's local impact. “It takes decades to build something like that but, evidently, only 18 months to dismantle it,” says Stephen Forrest, vice-president for research at the University of Michigan.

Pfizer is facing shrinking profits, expiring patents and a pipeline that is looking short of obvious blockbuster drugs. Late last year, it was forced to withdraw torcetrapib — a cholesterol drug that had been touted as a successor to Lipitor — from clinical trials (see Nature 445, 13; 2007).

LaMattina claims the move is needed because the company has so many drug candidates at the expensive, late stages of development all at once, yet needs to control costs. “We're a little bit a victim of our own success,” he says.

Pfizer will stop some early-stage work, including efforts to find new skin and digestive-tract drugs, and will cut several layers of middle management. Afterwards, LaMattina adds, its laboratories will work at 95% of capacity instead of 65%. Research overhead costs will fall by one-fifth, he says, freeing up hundreds of millions of dollars from bricks and mortar to be spent on science.

The four research sites after the reorganization will be at Groton, Connecticut; Sandwich, UK; St Louis, Missouri; and La Jolla, California. Consolidation into disease-specific groups at these sites will result in “fewer frequent-flier miles and more time at the bench doing science”, LaMattina says. The company is planning a new emphasis on bioetherapeutics at St Louis, on cardiovascular work at Groton, and on vaccines at Sandwich and La Jolla.

But Alan Saltiel, director of the Life Sciences Institute at the University of Michigan and a former cell biologist at the Ann Arbor laboratory when it belonged to Warner-Lambert, before it was acquired by Pfizer in 2000, thinks the separation by disease area has a downside. “What they lose are the opportunities and synergies across therapeutic areas,” he points out. “A lot of drug discovery is serendipity.”

As well as closing the Ann Arbor site and two smaller Michigan research groups, the company will shut its research facilities in Nagoya in Japan and Amboise in France.

LaMattina says that the company is hoping to transfer, rather than dismiss, up to 70% of workers at targeted facilities. But Tony Butler, a pharmaceuticals analyst with Lehman Brothers in New York, doubts whether as many as that will move.

And Peter Rost, a former vice-president of marketing and strident critic of Pfizer's current management — now in litigation with the company over the circumstances of his departure in 2005 — predicts that there is worse to come. “This is just the beginning,” he says. “It is not the bottom. Two years from now, Pfizer will make another announcement, and cut another $2 billion. Just watch.”