Eli Lilly's Drug Assembly Line

Shortly after becoming CEO of Eli Lilly (LLY) in April 2008, John Lechleiter, a former lab scientist, sent senior executives a gift. It was a digital clock counting down, second by second, to Oct. 23, 2011. That's the day Lilly's $5 billion-a-year schizophrenia pill, Zyprexa, goes off patent. Next to the countdown were four words: "Do what we do."

The staff got the message. Lilly stands to lose $10 billion in annual revenues between now and the end of 2016—almost half its 2009 sales—as patents on three key drugs expire. To replace that, the company must pick up the pace of drug development.

The top six U.S. drugmakers all face such "patent cliffs," but Lilly's is one of the steepest. Making things worse, two of its promising experimental drugs failed last year in late-stage trials. Lilly's shares fell 11% in 2009, making it the worst performer among the 11 U.S. drug companies in the Standard & Poor's Pharmaceuticals Index. "Investors seem to price Lilly's stock as if it will never develop another drug," says Michael P. Krensavage, manager of Krensavage Partners, a hedge fund that owns Lilly shares.

When the company has faced such challenges in the past, "each and every time the answer has been new, innovative products," says Lechleiter, whose mission is to introduce two products a year starting in 2013. To reach that goal he has radically restructured the company, dedicating two out of five new business units specifically to developing drugs for cancer and diabetes. Rivals such as Pfizer (PFE) and Johnson & Johnson (JNJ) have also refocused their research efforts around specific therapeutic targets. But there's a difference: Lilly hopes to turn drug development into something akin to automobile manufacturing—an assembly line of tasks that work in concert to make the final product quickly and efficiently.

As a first step, Lechleiter in September placed every department that plays a role in turning molecules into medicines under one roof, from the people who assess side effects to the folks who deal with the Food & Drug Administration. One goal at this new outfit, called the Development Center of Excellence, is to take a standard three-phase drug testing process and compress it to two for certain products.

The idea proved its value last year. Instead of completing one trial to determine the best dose for a diabetes drug, then launching a separate trial to determine efficacy, Lilly tested several doses at once. A team of statisticians continuously assessed the data, allowing researchers to eliminate dosages that weren't working. Once the optimal dose was known, researchers continued to test it on the same group of patients, rather than recruiting new ones for a separate phase. Timothy J. Garnett, Lilly's chief medical officer, says this approach could save the company 14 months.


The company is also adopting a technique known in industrial manufacturing as "critical chain." Say a team of people ordinarily is given three weeks to produce a 30-page report on a clinical trial. The critical chain process assumes precious time is lost as team members responsible for writing, publishing, and signing off on the report try to multitask, says Garnett. Tell each one exactly when he or she must complete a particular task, and three weeks may shrink to just seven days. "If we say, 'you only have a week, but that's your No. 1 priority,' it will only take a week. There's a self-fulfilling element to it," Garnett says.

Some experts worry critical chain will be a bad fit with drugmaking. The process of developing a new medicine is fraught with uncertainties, and time lines can be thrown off-kilter by external forces, such as delays at the FDA. Assembly-line processes "work better in jobs that are predictable"—unlike the pharmaceutical business, says Richard "Erik" M. Gordon, assistant professor at the University of Michigan's Ross School of Business.

Lechleiter's obsession with efficiency dates back to the 1970s, when he pursued his PhD in organic chemistry at Harvard University. His lab focused on reducing the number of steps needed to make new compounds—an experience he found useful when he joined Lilly's agricultural division straight out of graduate school, working on herbicides. One of his first challenges was to devise a more cost-effective way to make a particular molecule. "He solved that problem and simultaneously developed an enabling technology that allowed others at Lilly to progress with their research," recalls Lechleiter's mentor and former PhD adviser, Paul A. Wender, now a chemistry professor at Stanford University.

Since becoming CEO, Lechleiter has emerged as a communicator within the company. He maintains two blogs on Lilly's intranet and attends Wednesday morning meetings where Lilly scientists present their work. Although it has been more than two decades since he sat at a lab bench, he says he's not afraid to ask a question or two. "Usually I preface it by saying something like, 'Excuse me if I'm really ignorant,' " Lechleiter says.

These days, it's Wall Street types, not complex chemistry, that hand Lechleiter his toughest challenges. Lilly spent 21% of revenues on R&D in the most recent quarter—far outpacing the industry's average R&D outlay, which is closer to the midteens. "A lot of investors say: 'Unless you cut back, you may be spending 24% of sales on R&D,' " says Lechleiter. "I say: 'So what? That's what we're here to do.'"

Barclays Capital (BCS) analyst Tony Butler says the impact of the efficiency-focused initiatives is unclear: "It's clever, but how much money will it save? We need to see the fruits of research in risky areas like Alzheimer's."

Lechleiter has yet to deliver the results Wall Street craves. Last summer, Lilly shelved a multiple sclerosis drug and an osteoporosis treatment when they came up short in clinical trials. And the new blood-thinning drug, Effient, has made little headway against Bristol-Myers Squibb's (BMY) $6 billion-a-year Plavix. Meanwhile, the turnaround clock keeps ticking: Lechleiter has about 600 days to prove his strategy can protect Lilly when it's far out on the edge of the patent cliff.

    Before it's here, it's on the Bloomberg Terminal.