Sidney Taurel is writing the closing chapter of his 34-year career at Eli Lilly (LLY
). On Oct. 1 the 56-year-old chairman and chief executive handed over his day-to-day responsibilities to John C. Lechleiter, Lilly's new president and chief operating officer.
Lilly is going through a transition as well. Taurel sees the $13.86 billion company evolving from an outfit that discovers and markets drugs for vast populations to one that targets specific subgroups of patients based on their distinct genetic makeup.
Taurel cautions that he cannot foresee how long this metamorphosis to personalized medicine will take. However, he says it's inevitable, given trends in science and the marketplace.
WRINGING OUT COSTS. In the meantime, Taurel says, Lilly must squeeze out greater profits from the old way of doing business. That means slashing the costs of bringing a compound from lab to patient. It also means hiking sales from existing drugs that have been bruised by worrisome side effects.
These include Lilly's No. 1 seller, antipsychotic Zyprexa, which has been linked to substantial weight gain and diabetes, and antidepressant Cymbalta, which the Food & Drug Administration just concluded should be off-limits to patients with liver problems because the drug could hurt them further.
Sitting in an easy chair in his 12th floor office in Lilly's Indianapolis headquarters before hosting an employee luncheon, Taurel talked at length with BusinessWeek Senior Correspondent Michael Arndt about how his company is preparing for the future of medicine. An edited transcript of the conversation follows:
Where do you see the pharmaceutical industry heading?
There is still a very, very large number of diseases today for which we don't have a solution. Think about Alzheimer's or all kinds of cancer or stroke. Going forward, we see a need to improve the value proposition for society, for patients, for providers, for payers.
The current value proposition is a one-size-fits-all model. We're going to providers and saying: Try this drug on patients who have this disease and see what happens.
In the best of cases, 80% of the patients will respond to the drug positively, and maybe one out of 1,000 will be hurt by it. In the worst of cases, only 20% to 40% will respond positively, and it may have side effects on one out of 100 or 50 patients.
What if we could go to the providers and the payers and the patients, and basically say: You can have 90% or 100% probability that you will have a positive outcome in that population of patients, and don't use the drug at all in this other population of patients? That's the long-term vision of personalized medicine.
Does that work economically for the company? With a blockbuster, you get the biggest bang for your buck, right?
In many cases, it probably will mean lower sales, because the targeted populations will be smaller. But in some cases, it may mean higher sales.
If you have a very good value proposition, the majority, if not the totality, of the patients who have that particular condition will adopt that product, rather than 5% or 10% of the patients. The pool may become smaller, but it may become deeper.
Let me give an example where this would be so useful: Zyprexa. The National Institute of Mental Health just released a study that compared all of the antipsychotics. Zyprexa came out best on how long people stayed on a drug.
At the same time, though, this study confirmed that for a number of patients, maybe up to 30% or so, there is significant weight gain. If we could predict in the future who are those 30% with a biomarker or some other tool, I'm pretty confident the product would be used more broadly.
Are you looking for that biomarker?
We are obviously looking very hard at how we can do that, but it will take time. The challenge as we move forward is to make the current model much more cost-efficient.
Our expectation is that biomarkers will help us lower the cost of developing drugs. Instead of a shotgun approach, you're going to have a very targeted approach, and select ahead of time those patients who are most likely to respond. We hope this will result in smaller clinical trials and lower attrition -- the failure rate of drugs, which is the biggest cost today.
Since you haven't yet discovered many of these biomarkers, what can you do today to reduce costs?
Right now we estimate that the average cost of bringing a drug to market is $1.1 billion; this includes all the expense of all the drugs that are killed before they make it to market. If we continue business as usual, this figure will go up to $1.5 billion or more by the end of the decade.
We are working on a number of projects to bring that cost down to $800 million. These include trying to kill projects earlier before we've spent a lot of money on them. Here again, biomarkers would be helpful.
We are also doing more clinical trials in countries outside the U.S. with lower costs. In India and China, we are starting to see some centers of excellence with very, very good clinical trial work. Also, we are outsourcing more of our work in chemistry. Today, 20% of all of our chemistry work is done in China, at about four times less than the cost elsewhere.
I want to ask you more about Zyprexa. In the second quarter, Lilly posted a loss after taking a $1 billion charge to settle thousands of lawsuits by patients who claimed that Zyprexa had caused not only weight gain but diabetes. But while announcing this charge, you also said that these claims were invalid. Why are you offering a settlement if you think you are right?
That's a sad commentary on the state of our tort system. Whether you are going to win or lose in the courts, there will be a lot of publicity around each trial.
The patient taking this product is, by definition, a vulnerable patient. We've seen the very deleterious effect that trial lawyers advertising to recruit patients has had. Some patients have come off their medication, and some doctors are now afraid of prescribing it. And so we decided to settle to avoid all those effects that would be bad for patients and a big distraction for the company.
Do you think this tactic will work?
We have had a fairly positive experience outside the U.S. Two or three years ago in Japan, there was a prominent warning on the risk of diabetes. As physicians became aware of how to manage the issue, sales resumed their growth.
That is what we are trying to do in the U.S. Sales of Zyprexa have been declining here, but we think the erosion has slowed down. Obviously, time will tell.
In addition to mental-health drugs, Lilly is big in diabetes medications. You could argue that treating diabetes with drugs is the good and right thing to do. On the other hand, lifestyle can eliminate or control diabetes. So by having drugs, aren't you giving people a crutch, enabling them to maintain bad lifestyles with poor diets or no exercise?
That would be a mistake. We're all about helping people having better health care. If we are to be accepted by society, it is important that we simply not be pushing our pills.
In the case of diabetes, you are right: Many people can control their diabetes by diet and exercise. That would be the first line of attack. But we know that in a lot of people their disease will progress to the point where they need a product, and then they have to go to insulin.
Along those lines, there are critics of the industry who say drug companies sometimes almost concoct a disease because they've got a drug that would fix it. Erectile dysfunction might be one example of that.
I will not sit here and argue that an ED drug has the same medical impact of a drug like Xigris that can save lives of people with severe sepsis. But there is still a medical need.
Any last thoughts?
We are very focused to make sure we execute on our strategy, and that's why we have recently made John Lechleiter president and COO, to ensure that I can focus more on long-term strategy and policy issues while he focuses on the implementation day to day.
Does that mean you're going to be stepping down soon?
No, it means that we're going to be hopefully doing a better job than we've done so far in both developing and implementing strategy.