CEO Interview: Novartis' Dan Vasella

Novartis has done better than most Pharmas over the past five years, though shareholders are still dissatisfied. But CEO Dan Vasella continues playing for the long-term -- pursuing R&D innovation and cost-effective dealmaking, all within a relatively conservative strategic framework.

An aggressive dealmaker and strategist, Dan Vasella is also hedging his company’s long-term bets.

By Roger Longman

All in all, it’s been a rough start to 2007 for Novartis AG’s flagship pharmaceutical division and the company’s chairman and CEO Daniel Vasella, MD.

In February, the Food and Drug Administration rejected the company’s most important pipeline project, vildagliptin (Galvus), which had not long before promised to be a first-in-class medicine worth billions in sales. Instead, Novartis had to watch Merck & Co. Inc. —whose program started years after Novartis’—zoom ahead, win approval, and, perhaps, an insuperable lead in this market. (See "Januvia: Defining Primary-Care Success and Risk," IN VIVO, March 2007 [A# 2007800045].)

In early March, with the Galvus news and last year’s tepid stock performance fresh in the minds of investors, shareholders at Novartis’ annual meeting complained about Vasella’s pay package and his twin roles as chairman and CEO. Vasella was easily reelected to his jobs, but the company scrapped his golden parachute (three times his annual compensation).

Then, at month’s end, the FDA told Novartis to stop selling, temporarily, at least, its irritable bowel syndrome drug tegaserod (Zelnorm)—a blow worth $500 million in annual sales. According to sources close to the company, Novartis had argued that the drug’s use should indeed be restricted, but only to populations at risk for the cardiovascular events that had been uncovered. FDA apparently disagreed. Indeed, the temporary withdrawal represents, economically speaking, the worst of both worlds: if the drug comes back on the market, it will undoubtedly do so with a strict monitoring program, which will probably cost the company nearly as much as it will get in sales from its now-reduced market.

Admittedly, all that’s a glass-half-empty view. Novartis has actually done rather well with approvals, with three US go-aheads in the last six months: two anti-hypertensives--aliskiren (Tekturna) and Exforge (a combination of two older anti-hypertensives, Novartis’ own valsartan [Diovan] and Pfizer Inc.’s amlodipine [Norvasc]), plus the hepatitis drug telbivudine (Tyzeka) from its majority-owned affiliate Idenix Pharmaceuticals Inc. And it doesn’t have the big patent-expiration problems of some of its major competitors. In 2007 it will lose some portion of the $1.7 billion in revenues it derives from patent-expiring nail-fungus drug terbinafine (Lamisil) and anti-epileptic oxcarbazepine (Trileptal), but it doesn’t really have to worry about a major hole in its product lineup until 2012, when the patent runs out on the $4.2 billion Diovan.

The verdict is out on its new products. The most important is probably the renin inhibitor Tekturna, a drug Novartis had basically abandoned because of its formulation challenges but which was revived by a Novartis spin-off, Speedel Group , which solved some of the problems of making it. And if most analysts were surprised by FDA’s thumbs-down on Galvus, they were equally surprised by the agency’s thumbs-up on Tekturna, whose dosing remains far from ideal. Only a tiny fraction of the approved dose gets absorbed into the blood stream, but higher doses were nixed for side effects. Analysts are mixed on its prospects in a crowded hypertension market.

But there’s no question that Tekturna is an innovative product--a first-in-class success among a slew of same-class failures. Indeed, Novartis under Vasella has focused more than most on scientific innovation as a fundamental strategy.

That’s the reason behind its unusual R&D structure. Vasella has kept control of clinical development in Basel, where it’s run by a Novartis veteran, James Shannon, MD, who reports to Thomas Ebeling, the head of Novartis’ pharmaceutical business, who in turn reports to Vasella. But Vasella moved discovery to Cambridge, MA, putting it under the command of a well-known Harvard research physician but industry novice, Mark Fishman, MD--who reports with no intermediary boss directly to Vasella. Fishman was given plenty of money and latitude, which he’s used to build a huge new facility, hire an additional 1500 scientists, and create a new discovery strategy for the company, one that’s still unusual—at least to the extent Fishman pursues it—in the industry.

His notion: pursue human proof-of-concept (POC) trials in what might be called subset diseases. They'll attack broader medical problems by first going after more easily testable, phenotypically related conditions, preferably those caused by a clear genetic mutation in the same pathway. The theory is that when the drug proves itself, and relatively quickly, in the model disease, it will also be more likely to work in the broader condition. Attrition rates will drop; the number of innovative medicines will increase. (See "Novartis’s Research Experiment," IN VIVO, May 2006 (Also see "Novartis's Research Experiment" - In Vivo, 1 May, 2006.).)

According to Vasella, his R&D experiment is succeeding--a steadily increasing number of compounds are entering early-stage clinical trials (See Exhibit 1). But few if any of these drugs have yet moved into the most visible part of the pipeline--Phase III, which remains relatively weak.

Indeed, with all the advantages of 20-20 hindsight, the recent drug development programs that have worked out best for Novartis are those around less innovative drugs--like Exforge--or those that are done, in part or in whole, by outsiders--like Tekturna, ranibizumab (Lucentis), in-licensed from Genentech Inc. [See Deal], and Tyzeka. Whatever the intrinsic molecular problems that tripped up Galvus—an unquestionably innovative Novartis program—insiders say the company’s developers simply became complacent, unwilling to see the increasingly dangerous competition coming from Merck, which—recognizing that the first to launch among the relatively undifferentiated compounds in the class would likely have an enduring commercial advantage--streamlined its development program to beat Novartis to the market.

Vasella’s bold moves to increase the focus on innovative R&D are complemented by his boldness—and cost-consciousness—as a dealmaker (See Exhibit 2). With nearly $14 billion worth of acquisitions since 2005, Novartis has been an aggressive shopper. (See "Novartis’ M&A Machine," IN VIVO, November 2006 (Also see "Novartis' M&A Machine" - In Vivo, 1 Nov, 2006.).) Others have shelled out more cash--Johnson & Johnson some $19.2 billion, the vast majority in its $16.6 billion buyout of Pfizer Inc.'s consumer business and the remainder largely in medical devices. [See Deal] And there have been some big one-off deals—such as Schering-Plough Corp. ’s acquisition of Akzo Nobel NV ’s Organon NV unit. [See Deal]

But Novartis under Vasella has been very focused on value-for-money. Thus, most of the acquisition money has gone to a business—generics—where Novartis has no competition from Big Pharma, and therefore where it doesn’t have to compete with the strategic premiums Big Pharmas can pay. Indeed, Novartis has been at the table—but it backed away from some of the biggest biotech acquisitions and alliances, unwilling to fork over the kinds of risk premiums other companies have. Likewise, its 2005 acquisition of Chiron Corp. provided it with a strategic benefit at a merely tactical price. [See Deal] Novartis, which already owned 42% of Chiron, was its only possible purchaser, and the share price of Chiron, beset by production problems with its flu vaccine supply, was low. But the acquisition put Novartis into the high-growth vaccine business, with the ability to take immediate advantage of the growing flu market and government concerns over pandemics—and therefore industry-friendly regulatory policies. Even Vasella’s acquisition in 2001 and 2003 of what is now about a third of Roche’s voting shares was a good deal—Roche’s stock has soared; indeed it has done far better than Novartis’. [See Deal]

What worries some analysts, and excites others, is that Vasella might go further, trying to do a Big Pharma merger. And Vasella doesn’t do much—as in the interview below—to dampen such speculation. With the markets awash in capital, it’s clear that Vasella understands he could finance such a merger by pre-selling the post-merger pieces he doesn’t want to major private-equity funds eager for a piece of the pharmaceutical business. Certainly, he doesn’t think Big Pharma M&A—as do many analysts, CEOs, and nearly all heads of R&D—is a waste of shareholder money.

But then again Vasella has a very long-term view of the business. Specialty pharmaceuticals will probably be Novartis’ most important profit drivers. The relatively new chelating agent deferasirox (Exjade), for example, is extraordinarily profitable—eventually a billion-dollar product believes Prudential Equity Group analyst Tim Anderson, MD, but because it requires only several dozen reps, equivalent in earnings to a $2 billion primary-care drug. Meanwhile, Novartis’ cancer program, led by imatinib (Gleevec), is forecast to grow at about 17% through 2008; its flagship cardiovascular/metabolism program at less than 15%. Vasella is reluctant to move too far toward specialty drugs because he sees the pendulum eventually swinging back to primary care, indeed pushed back by what he sees as the unsustainability of specialty pricing. Likewise, the generics business certainly can’t compete for profits with the pharmaceutical division—but Vasella is betting that his generics hedge will prove valuable as proprietary drug prices squeeze down. As with his Roche investment, he’s willing to bide his time.

Indeed, fundamentally, Vasella is a conservative. If the drug industry is changing, it’s not changing so radically, he believes, that it needs a complete change of strategic direction--toward specialty drugs, for example, or toward externalized research. Vasella, instead, continues to spread his bets. So far, it’s a strategy that’s worked well enough—Novartis is the number-three best-performing stock among Big Pharma over the past five years. But with a 6% annual increase in share price, it’s hardly a barn-burning performance. The real question is whether Vasella’s steady-as-she-goes strategy will ultimately return Novartis to the kind of performance shareholders with plenty of other investment choices require from it.

Q: What lessons have you learned from your recent primary-care approvals and delays?

The lesson of unpredictability. Sometimes you can prepare for it; sometimes you can’t. For example, our people said we wouldn’t be able to find a synthesis for producing Tekturna affordably. But Alice Huxley [CEO of Speedel] believed she could do it. She wanted [Speedel] to take it over entirely, but I said "no, only with a call-back option"—too often, especially with manufacturing, the things that people say are impossible become possible. So we co-financed the project, and it worked out well for everybody. The product is still very expensive to make, and we need relatively high doses to get the therapeutic level we need, so we still have a few challenges. But we were able to deal with the uncertainty in that case.

As for Galvus: I was less concerned about the approval than I should have been. In preclinical studies we saw skin side effects in monkeys. While we did not see similar side effects in humans, we observed peripheral edema at high doses above therapeutic levels. While there were very different reactions in monkeys and humans, there was concern from FDA that patients with, for example, renal insufficiency could experience skin side effects.

Q: What would you have done differently about Galvus?

In retrospect we should have done a study in patients with renal insufficiency. But at the time our specialists did not think that it was warranted. Otherwise, I don’t know what we could have done differently. As you do more, you find more. And once you’re on the market, you’ll see even more. But it also seems that other divisions at FDA may have taken a different course of action.

Q: Certainly the metabolic division [which reviewed Galvus] has been under some political pressure.

Yes, FDA is constantly being scrutinized, and this has consequences. People become more risk averse. I don’t blame them—it’s a very human reaction. But it is not always in the best interest of patients who may have to wait longer for a medication.

Q: Saying no to something is lower risk than saying yes.

Yes.

Q: Meanwhile, Exforge--which did get approved--probably had the least uncertainty about it.

Yes, we combined two of the most frequently used anti-hypertensives so a lot of the questions you get with novel drugs were not relevant. The interesting aspect of this novel combination is that one can use lower dosages of each, so patients are less likely to experience side effects, but they will benefit from enhanced efficacy--a real improvement. Especially for patients with relatively high blood pressure, this is important.

Q: But aren’t there lessons here about how to balance the risk in your portfolio? The lower-risk projects--the combination drug Exforge and Tekturna, which was de-risked by having someone else take over much of the development job—get approved. Galvus, a highly novel program Novartis did completely on its own, gets delayed. Does that tell you how you want to balance R&D and late-stage pipeline?

No, because you have to make these decisions early, when you know far less. When you have clear-cut problems with a molecule, you can get rid of it. But if it looks promising, you’ll pursue it, finding out more and more as you advance. Unfortunately that costs time and money. You also have to ask what the market and patients need--and whether you have anything in your portfolio that would fulfill these needs.

Of course, it’s great to have a compound from one’s own discovery efforts. It creates great satisfaction, and you don’t have to pay licensing fees. But I am pragmatic. In the end, the most important question is whether we can provide a better treatment.

Secondly, things don’t always end up where they started. We’ve got a cancer drug--Zometa--that we’re developing as an osteoporosis drug--Aclasta. It’s better to remain open and interested--and to encourage researchers. At the same time you have to be willing to accept occasional setbacks.

Q: But doesn’t the Zometa example suggest that you should be focusing more on specialist drugs--where the trials are faster and smaller and the toleration of side effects is greater? Wouldn’t that lower the risk you’ve experienced with primary-care products such as Galvus?

It is wrong to either focus entirely on one or the other category unless you are seriously resource-constrained. Not too long ago, people wanted to stop working on specialty products. They said "look what’s happening in the industry, at Pfizer, at Merck; you have to focus on GP products and forget specialty products." Now the same people are saying that one should just be in the specialty businesses. At Novartis, we formed a GP unit and a specialty unit—that was sometime in 2001. And we’ve given the proper focus and attention to each.

The fact is we are large enough to do both. And there is an ongoing need for good GP products. Sure, approvals now require significantly larger studies; you need much more comprehensive marketing and selling efforts to be competitive. But depending on specialty drugs has its own risks, given the prices companies have to charge when selling drugs for small numbers of patients. The increase in prices has continued to accelerate. First, cancer therapies were priced at $50,000 a year; then $100,000. With combination therapies you can reach $150,000, and next we will see prices of $200,000 or more. It’s just not a sustainable model. Of course, we need certain price levels to keep the field attractive for investment, but at the same time we have to be circumspect in setting prices.

Q: And yet Novartis’ strategy for early-stage development focuses on very small "model" diseases, not big primary-care indications.

That’s true, but if a drug’s mechanism of action targets a certain disease extremely well, even if that disease is small, it’s highly relevant to the patients who are suffering. It is equally important to have a hypothesis regarding how the mechanism works, and to have a reliable human model. If you have a product with several indications you can build on these. Who knows what a medicine could develop into? Some people know everything years in advance, but drug developers don’t.

Q: Five years ago, you restructured R&D—moving discovery headquarters to Massachusetts, with a new chief, an academic physician, Mark Fishman, reporting to you. Development, headed by a Novartis veteran, stayed in Basel, reporting to the president of the pharma division. You’ve had time to assess the move: did it work? And what didn’t work?

Yes it worked. You can tell that by the number of compounds we’re discovering and advancing into our pipeline, and by the fact that we can now attract people of the highest scientific standing, who we were not able to attract before. Good scientists attract good scientists, and poor scientists attract poor scientists. And good science is an absolutely necessary foundation for good discovery and for finding novel solutions for patients. Not sufficient, but essential.

But clearly there were risks to our approach. If there had been personality difficulties between Mark Fishman and Joerg Reinhart [at the time of the reorganization, head of development, now head of Novartis’ vaccine business], that would have hurt. But in fact Joerg and Mark worked well together.

Mark came straight from academia and so had to adapt to a steep learning curve, which he did. Despite all modern communication tools, geographic distance remains a challenge—with research in Cambridge and development in Basel. But it is workable.

Q: Is there something the R&D group has not delivered that you’d expected them to deliver?

No, not for the time being. Of course, there are always some aspects of R&D that aren’t fully satisfying--and some areas that you think are doing great. But targets are moving. What’s great today is insufficient tomorrow. I’d like to have even more new compounds coming out of the Novartis Institutes for BioMedical Research --and it has been increasing every year. But what was fine last year is not fine for this year, and will be even less so next year. It’s really the work of Sisyphus: it’s never ending. We’ll never be completely satisfied.

Q: Does the buildup in Cambridge make you less willing to look outside for the best science?

No. It’s allowed us to improve internally, that’s true. Biologics is now flourishing at Novartis because Mark has been pushing it--and at the same time bringing a lot of the technology in from outside, for example in terms of RNAi. There are two ways you stop looking externally: if you’re not good enough scientifically, you may become defensive about what comes from the outside, as one may experience it as the proof of failure; and then there are companies that have reached the pinnacle and believe no one is as good they are--and they don’t consider external opportunities. One has to be constantly thirsty for new discoveries, both inside and outside the company. We are now setting up a discovery operation in China--real discovery--and thereby we will tap new talents to foster our innovation power.

Q: Activist shareholders have become a force in Big Pharma. They certainly helped push out Gilmartin from Merck, McKinnell from Pfizer, and Dolan from Bristol-Myers Squibb. And you’ve had your own problems with them. Are these activists a good thing for the industry?

Sometimes, perhaps. We’ve seen an evolution of power from the CEO to the board, and now from the board to activist shareholders, who are generally not owners, incidentally, but agents. But do these shareholders have the information they need, in time, to make good decisions? What are their real interests and time frames? These shareholders generally don’t know the detailed objectives agreed between the board and the executives and whether those objectives—sometimes important long-term objectives—are being met or missed. And the consequence could be—could be—that CEOs become more short-term oriented and more political, engaging in actions that please shareholders whether it’s in the best long-term interest of the company or not.

There are so many stakeholders: shareholders, employees, customers, politicians, regulators, nongovernmental organizations, all with their own agendas and incompatible expectations. Boards and CEOs are facing an increasingly challenging responsibility. It also costs effort, money, time, and mental energy. No wonder that so many turn toward private equity.

Q: Do you expect further M&A within Big Pharma?

Yes, I think we’ll see some. There will be continuing pressure on margins; and that will lead, almost inevitably, to more industry consolidation. Transactions of course involve risks and hard work, but for some companies it is still worth it.

Q: But the Big Pharma consolidation of the past hasn’t done much for the industry.

You don’t know what those merged companies would have looked like without it. It’s an uncontrolled study with very few subjects. The success or failure of these deals has often been defined after the fact, not prospectively. You’d need to look at each of the companies’ pre-merger projections, including patent expiries and product launches, and then try to judge whether they’ve done better, after the merger, than one could have expected them to do alone.

Q: And Big Pharma M&A wouldn’t be stymied by anti-trust concerns?

Not in general. Sure there are specific examples of companies with major overlaps in specific important product areas. And when there are too many overlaps, and the anti-trust authorities force divestitures, a combination doesn’t make any economic sense.

Q: Private equity is playing a bigger role. How will it affect the industry?

Their role is increasing because the private equity firms are sitting on a lot of cash and would like to find good investment opportunities. The problem is that as long as strategic buyers are willing to pay significant premiums because they can realize synergies, private equity will find pharmaceutical prices unattractive.

Q: But one could assume that if you bought another big drug company, you could help pay for that by using private equity to absorb the things you would like to spin off.

Absolutely.

Q: Could you work together, up front, with a private equity firm to go after a target?

Oh sure--perfectly conceivable. It’s, however, not easy striking such agreements up front. But if each player clearly sees the assets it wants and what it doesn’t want, yes, of course.

Q: Novartis is the only Big Pharma that has built a significant generics business. Why?

It’s profitable; it’s growing; we’re serving our customers and are competitive: what more do you want? Generics are going to be a fundamentally more important part of the pharmaceutical industry—no doubt. The prescription volumes [of generics] will continue to grow because of demographics and cost containment.

Q: Has there been any managerial or tactical benefit for the pharma business from Sandoz [Novartis’s generics unit]?

We’ve learned something about the vulnerabilities of patents. Sandoz produces pharma’s fermentation-based products. But there are basic differences in how these businesses work. And I don’t think our generics business has had much effect on how we manage our pharma business.

Q: What’s your short- and long-term vision for the vaccines business?

Near term, the opportunities are in flu and meningitis. At our clinical trial in New Zealand, for example, the death rate dramatically dropped after the introduction of our meningitis vaccine, albeit for a specific strain of meningitis. Longer term, we’ve initiated projects in areas where Chiron already had ideas--but which it couldn’t fund when its vaccines projects had to compete with its biopharmaceutical projects. I can’t disclose the ideas for competitive reasons, but they’re good targets with potentially great health impact. Likewise, Chiron couldn’t really afford geographic expansion, which is another big opportunity for vaccines. Governments are now more willing to open up [their vaccine markets] because they recognize that with the potential for pandemics they’ll need more supply than they can hope to get from local manufacturers.

Q: And so not only will they be more flexible from a regulatory point of view, they’ll be willing to pay higher-than-historical prices for vaccines?

Only to a limited degree. But yes, the introduction of Gardasil shows how pricing could change.

Q: You’ve certainly generated a lot of controversy with your response to India’s decision to allow generic Gleevec. Does this issue really merit the very public stance you’ve taken—and the criticism it’s generated?

All we are really saying is this: either you have a credible patent law compatible with WTO rules or you don’t. If you don’t have one, please tell us; then we’ll know how to behave. If India wants to attract biomedical R&D it has to provide appropriate protection. Look, there are few things as crucial to the future of innovation as IP protection. If governments undermine that, then they undermine the essence of what it will take for the industry to be successful in the long term and to be able to provide new and better medicines for patients. The trend unfortunately is not positive. Legislators, society, and the press should be cognizant that industry has delivered; that many diseases, which once couldn’t be dealt with, are now manageable or even curable because of what we’ve discovered and developed. To endanger that medical progress would be a terrible mistake. What the problem in India shows is that we’re sometimes going in the wrong direction. The US cannot finance medical innovation for other countries; there has to be a fair sharing of the costs. Every country, to the degree it can, should contribute.