Novartis: Radically Remaking Its Drug Business

Five neuroscientists file nervously into Novartis' (NVS) vast boardroom in Basel. CEO Daniel L. Vasella is seated at the table, along with nine senior executives that make up the company's innovation board. The scientists hope to persuade Vasella to move forward on an experimental vaccine for Alzheimer's disease, an illness that afflicts 25 million people worldwide. Even a moderately effective treatment would be a blockbuster, but testing the vaccine would take years and several hundred million dollars. When one of the scientists points out that other drugmakers are targeting the illness, Vasella just shakes his head. "We still don't know whether Alzheimer's is one disease or several, or if the current approaches even address the right disease mechanisms," he says, frustration edging into his voice. He sends the proposal back for a rethink. Vasella's ability to quickly set aside one of the world's most troubling diseases reflects a radical shift for Novartis—and one that may offer the drug industry its best chance to come up with the kinds of innovations that have eluded it for so long. Most big drugmakers shower their research and development funds on diseases such as cancer or depression, where huge potential markets beckon despite a deficit of scientific understanding. In recent years this approach has led to high rates of failure when drugs are tested in clinical trials. Seven years ago, Vasella, 55, took a calculated risk and turned Novartis' R&D model upside down. He resolved to push drugs through the long testing process only if they are backed by proven science. It doesn't matter whether the diseases the drugs treat are rare and the initial markets minuscule. Once a drug proves its worth against one disease, Vasella reasons, it can be tested against others. "If you are guided purely by financial estimates and not the science, you end up wasting time and money," he says. Vasella's goal is to institutionalize the lessons from one of the company's most successful creations, the cancer drug Gleevec. Initially approved for a rare blood cancer that strikes just a few thousand people each year, it has proven effective against six other life-threatening diseases. The drug pulled in $3.7 billion in revenues last year for Novartis, which posted a total of $9 billion in operating income on sales of $41.5 billion. Vasella admits Gleevec was something of a fluke: The drug has a complicated history that mostly predates Novartis' shift in strategy. Yet Gleevec's triumph helped crystallize Vasella's thinking. Today Novartis has 93 drug candidates in the pipeline, 40% more than three years ago, and 80% of Novartis' drugs last year made it from early testing to late-stage development. That's a 60% improvement over 2005.


If Novartis can quickly build on this track record, it could mark a turning point for both the company and the industry. While plenty of biotech companies place a premium on science, for the past two decades most pharmaceutical companies have been more interested in developing and marketing blockbusters aimed at major diseases. In years past, this approach produced giant wins, including a host of famous cholesterol drugs as well as other moderately effective, mass-market medicines. But experts say drug companies have exhausted the easy targets. With patents on many older blockbusters starting to expire, the industry is poised to lose an estimated $140 billion in sales to generic competition over the next five years. Those revenue sources must be replaced. Despite multibillion-dollar research budgets, none of the top companies has a wealth of promising compounds in its development pipeline. The industry also faces regulators more vigilant than ever about safety, and health insurers starting to balk at covering costly drugs that bring only modest benefits. All these travails help explain the recent frenzy of drug-industry mergers and acquisitions, including the $68 billion union of Pfizer (PFE) and Wyeth (WYE) and the $41 billion marriage of Merck (MRK) and Schering-Plough. Yet the bigger these companies become, the fewer new drugs they seem to produce. U.S. pharmaceutical makers spent $65.2 billion on R&D in 2008, nearly double the tally in 2002, but only eight completely new or first-in-class drugs reached the market last year, half as many as in 2001. A former physician, Vasella reckons drug executives have strayed too far from their scientific roots and forgotten that their core customers are patients, not shareholders. "It's the tyranny of quarterly earnings," he says, a trace of an accent revealing his Swiss German upbringing. Vasella believes that focusing Novartis' research on smaller, narrowly defined groups of patients will lead to more targeted and effective therapies with fewer side effects. Regulators will therefore be inclined to approve them, and reimbursement will be less troublesome. "With so much waste in the health-care system, insurers cannot afford to bankroll me-too drugs," Dr. Lee N. Newcomer, senior vice-president for oncology at United Healthcare (UNH). "But if new approaches such as Novartis' mean the right patients will get the right drugs, insurers will likely pay for them." Vasella's vow to break with the status quo has won plaudits from management gurus who marvel at the inefficiencies in the drug sector. "He is not a prisoner of the existing paradigm. Instead, he is trying to change it," says Ram Charan, a leading consultant and author. "Vasella is able to anticipate and visualize radical change much the same way Steve Jobs did at Apple (AAPL), Andy Grove did at Intel (INTC), and Sam Walton did at Wal-Mart (WMT)." William W. George, professor of management practice at Harvard Business School and the former chairman and CEO of medical device maker Medtronic (MDT), says Vasella "has the mind of a long-term strategist." Some on Novartis' board were skeptical at first. "They kept asking: 'How will we know this huge investment will work?' " recalls George, who resigned from the board this past February when his son Jeff became head of Novartis' generics business, Sandoz. "Dan said: 'You won't know for at least a decade.' " The remark was vintage Vasella: confident, direct, and searingly honest. When I first met him nearly a decade ago, Novartis had stakes in both agribusiness and pharmaceuticals. Vasella believed gene-based sciences would soon prove the value of this union, but the synergies never materialized. Where hubris might have driven a less self-aware CEO to throw good money after bad, Vasella quickly ditched the seeds and chemicals business and focused on drugs. It was a natural decision, given Vasella's unusual status as a physician-CEO in a profession where nearly all top executives have backgrounds in law, accounting, or marketing. "The industry really needs more leaders who are as passionate as [he] is about the business. He is a CEO who thinks like a doctor," says George. Vasella's ability to keep patients in mind may come from his experiences not only as a physician but also as a patient. He described these to me in 2003 when we met to talk about Novartis' hopes for Gleevec. As a five-year-old growing up in the Swiss town of Fribourg, Vasella developed asthma and spent two summers on a farm in the mountains, separated from his family. At eight he contracted tuberculosis, followed by meningitis, and spent a year in a hospital and a sanatorium. But his greatest sorrows came later, starting when he was 10. "My older sister had Hodgkin's lymphoma," he told me. "I watched her over three years as she slowly got weaker. She died at 19." Three years later, his father died in surgery. A second sister, also a doctor, was killed in a car crash. Perhaps because of these tragedies, Vasella is devoted to his family. At the recent innovation board meeting, I was impressed when he unapologetically took a call from one of his three children.


A childhood blighted by illness and loss led Vasella to a career in medicine. As a resident at the University of Bern, he worked with Dr. Rolf Adler, then head of internal medicine, who encouraged him to undergo psychoanalysis. It helped him work through his losses and sparked a lifelong interest in the field. (Vasella collects first-edition works of Sigmund Freud, among other authors.) Analysis also made Vasella a better doctor, says Adler, now professor emeritus at the University of Bern. "He was always question- ing both himself and others, but he was a good listener, too." Adler was disappointed when the young doctor decided to leave medicine for a career in business in 1988. The reasons were telling: Vasella was unwilling to wait out the years it would take in academia to be given a chance to lead. Moving to New Jersey, he took a position as a salesman for the Swiss drug company Sandoz, where his wife's uncle was chairman. In 1992 he returned to Switzerland to head marketing, and two years later he took over as CEO. In 1996, at the age of 43, he pulled off a merger between Sandoz and another mid-tier drugmaker, Ciba-Geigy, to form Novartis. Vasella's boldest move came in 2002, when he abandoned the traditional drug-development model. He declared that Novartis would investigate only diseases for which new drugs were desperately needed and where the genetics of the target illnesses were well understood. While other CEOs saw the pursuit of rare diseases as commercial suicide, Vasella believed many of the illnesses shared the genetic underpinnings with more common ailments. Vasella also decided to move Novartis' main global research operation from Basel to Cambridge, Mass., a short distance from MIT, Harvard, and other paragons of biological research. Vasella spent about $4 billion on the move and then quickly made another controversial decision: He recruited Dr. Mark C. Fishman, a renowned cardiologist from Harvard who had no industry experience, to run the center and overhaul Novartis' drug-discovery business. "When Dan first called, I thought he had the wrong number," says Fishman, whose work studying zebra fish resulted in the discovery of 100 genetic mutations involved in the cardiovascular system. Fishman was happy at Harvard and saw little reason to leave, but Vasella persevered. Fishman was the expert who convinced Vasella that medical research will reap the best results by focusing on a small number of important molecular pathways—the complex sequences of interactions among chemicals, proteins, and larger cell structures in the body that underlie all illnesses. "There are 24,000 genes in the genome, but there are only a few dozen pathways conserved throughout evolution," he says. With little known about how each gene functions in a larger cellular context, Fishman compares the genes to a mere list of words in a dictionary. Molecular pathways, he believes, are the missing grammar. Fishman reckons that finding all the links in a pathway and then locating the key signals that can turn genes on or off will lead Novartis to therapies for illnesses once deemed incurable: "The theory is, we will be able to apply them in disease after disease because the pathways are shared." While Fishman's new colleagues admired his knowledge of developmental biology, there were also hard feelings. Many Novartis scientists were struggling with the new management and the shift in power from Basel to Cambridge. "Some were angry and demotivated," Vasella concedes. "The changes bred a lot of insecurity." In particular, many questioned the choice of Fishman. "People in research wondered what value someone who had never led a research organization in the pharmaceutical industry brought," recalls James Shannon, chairman and CEO of San Francisco biotech Cerimon Pharmaceuticals and the former global head of development at Novartis. The biggest pushback came from the senior executives in sales and marketing who were used to calling the shots. Many couldn't fathom how a business model focused on small groups of patients would ever make money. The marketing side "was living happily off Diovan [Novartis' top-selling hypertension drug], only to have Mark Fishman tell them the age of the blockbuster is over," Shannon says. Things got worse when Fishman banned running commercial analyses of new drug candidates until the company had sufficient clinical data. This approach, backed by Vasella, was heresy in an industry that spends vast sums trying to assign a hypothetical value to each potential drug at every stage of the R&D process. "Dan threw a strategy on the table and said: 'This is it, go make it work,' " Shannon says.


The most important test case for Vasella's ideas may turn out to be an illness almost nobody has heard of. It's an inflammatory disorder called Muckle-Wells syndrome in which a single genetic mutation results in the buildup of proteins that can cause rashes, joint pain, and fatal kidney damage. Muckle-Wells afflicts just a few thousand people worldwide, too small a number to attract most drugmakers. But Fishman and Vasella have a hunch the drug they've developed for it will prove effective in more widespread diseases. "With other companies, rare diseases are an afterthought," says Fishman. "For us they are often the starting point." The drug, a synthetic antibody that can tamp down out-of-control immune responses, isn't new. Novartis briefly tested it against arthritis, then abandoned it. In 2004 one of Fishman's scientists asked to retest it in a trial involving just four patients suffering from Muckle-Wells. Within 24 hours of receiving the medicine, all four showed dramatic improvement, and within a week the disease was barely detectable in the blood. Patients remained in remission for as long as six months. Despite these promising results, development efforts soon bogged down in debates over what the drug's exact commercial value might be. "This is precisely what paralyzes the industry and leads to the destruction of innovation," says Trevor Mundel, a physician-cum-mathematician brought in by Fishman as head of global development. The confusion persisted for a year and a half, Vasella recalls with irritation. The initial tests "took way too long, much longer than necessary." Vasella ultimately resolved the disputes, and Novartis went on to test its drug in several rare autoimmune ailments. Using advanced computer models to show how different patients might respond to the treatment, Novartis was able to persuade the Food & Drug Administration that it was safe to move directly to late-stage trials involving hundreds of patients with Muckle-Wells and a related cluster of immune system diseases, thus shaving a year off the expected development time and saving tens of millions of dollars. While Pfizer, GlaxoSmithKline (GSK), and others are also using simulation, "Novartis is far ahead of the rest of the industry," says Dr. Howard Lee, director of the Center for Drug Development Science in Washington. Vasella places a high premium on medical experience. After a major restructuring in 2007 that led to the loss of 1,260 sales and marketing jobs, Vasella replaced some of the company's senior leaders with medically trained scientists who grasped his approach. According to Shannon, some of the old guard fled voluntarily. "They saw Daniel rewarding projects such as Muckle-Wells and said: 'If this is what the future of the company looks like, it's not for us,' " he explains. On June 3, The New England Journal of Medicine published important data on Novartis' phase 3 trial of the Muckle-Wells drug, now called Ilaris. More than 90% of children and adults suffering from the immune disorders the drug was designed to treat had rapid and sustained remissions. The drug is under priority review with the FDA, and Novartis hopes to seek approval for use against a variety of autoimmune diseases, including Type 2 diabetes and severe arthritis. There is much more in the pipeline. Novartis is neck-and-neck in a race with Germany's Merck to launch the first oral treatment for multiple sclerosis. And Novartis' drug Afinitor, approved by the FDA in March for advanced kidney cancer, shows promise in six other forms of the disease. In trial data released this month, Afinitor shrank tumors by 50% or more in one-third of patients suffering from lymphoma, a cancer of the lymph system. And Gleevec? Fresh evidence shows that it targets the same genetic mutation involved in certain forms of melanoma, the deadliest of skin cancers. The lesson, says Fishman: "Well-designed drugs just keep on giving."

Business Exchange: Read, save, and add content on BW's new Web 2.0 topic network

The Making of a Miracle Drug

Since its approval in 2001, Gleevec has turned a rare form of leukemia from a death sentence into a manageable disease. But with only an estimated 10,000 patients worldwide afflicted with chronic myelogenous leukemia, Novartis (NVS) initially had little interest in the drug. Dr. Brian J. Druker at Oregon Health & Science University's Knight Cancer Institute, whose lab did the basic work on Gleevec, persuaded Novartis to take a chance on it. In a 2003 Science Watch interview, Druker explains how he made the case.

For this and other stories on the drug industry, go to

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