Following Science

Sep 30, 2005

Outside Ian Clark’s window on DNA Way in South San Francisco, cement trucks rumble past, kicking up dust and gravel. Legions of construction workers pore over design plans, and, set off against the blue of the bay, mountains of lumber outline what will soon become a research center.

    “We give the scientists the best view,” says Clark, senior vice president of commercial operations at Genentech.

    Since its birth almost 30 years ago, Genentech has been uniquely devoted  to science, especially the science of recombinant DNA. Even ignoring for a moment the California trappings—executives pedaling bicycles to work, the tie-dye t-shirts and flip-flops on the research floor, the vegetarian burritos in the cafeteria—an aura of science surrounds the company. The corporate headquarters feels less like a business and more like a college campus—an even more impressive fact considering that Roche holds more than 55 percent of Genentech’s public stock.

    Nonetheless, Genentech is a business, and by many measures, one of the industry’s most successful: 2004 sales hit $3.75 billion, up 40 percent from 2003. Three quarters came from the oncology franchise.

    Genentech’s sales may not be enormous, but its enterprise value (market capitalization minus debt), rose more than 56 percent from 2003 to 2004, reflecting the market’s belief that its prospects are bright. Meanwhile, the company’s ratio of enterprise value to sales was the best among the industry’s top-16 publicly traded companies. (See Industry Audit,” Pharm Exec, September 2005.)

    The company’s growth is backed by an arsenal of innovative therapies and a stocked pipeline: Genentech is awaiting action or preparing the filing for some 10 new indications, while conducting Phase III testing of 14 new indications, including the highly anticipated Lucentis (ranibizumab), for age-related macular degeneration (AMD).

    “For Genentech, over the next few years, it’s basically an execution game,” says Frank DiLorenzo, biotech analyst for Standard & Poor’s.

    Clark will play a key role in making that execution happen. He first came to the company in 2003, leaving his post as president of Novartis Canada to become head of Genentech’s BioOncology group. At the time, that may have looked like a step backward. After all, Clark had spent the last five years in top leadership positions at Novartis. Now, here he was, running a franchise with only two marketed products to its name.

    Two years later, his move has proved to be a smart bet. Clark is taking on a new job as senior vice president of commercial operations, replacing high-profile industry veteran Myrtle Potter, who recently stepped down. Potter helped groom Genentech into the little biotech that could. Next, Clark, with his experience in oncology, must lay plans for the company’s new growth phase,

    If the first act of Genentech’s story was mostly about the science, act two will be focused on business. Under CEO Arthur Levinson’s direction, Genentech must work out new models for commercializing the company’s insights into molecular-level medicine. The company is pursuing a strategy that enables it to grow larger by targeting relatively small patient populations.

Avastin, the Platform Drug

Perhaps no product better illustrates Genentech’s approach to drugs than Avastin (bevacizumab), the first cancer drug to market to use the much-discussed strategy of antiangiogenesis—which some analysts project as the best-selling oncology drug of all time. In his office at Genentech headquarters, the man behind the drug, Napoleone Ferrara, slides his chair back from the 20-inch monitor of his I-Mac and offers the only seat not cluttered with paper, charts, or books. A Sicilian by birth, he is a modest man who can’t understand why his story still fascinates the public.

    “It’s a very old story by now,” he says. “It all began with the pituitary gland.”

    Ferrara came to the United States more than 20 years ago on fellowship to the University of California-San Francisco (UCSF). He was studying the pituitary gland to try to better understand what made blood vessels grow. Although the visionary Judah Folkman, among others, had already proposed that angiogenesis, the formation of new blood vessels, had clinical applications, no one had found direct evidence of the chemical mediators that control the process. Ferrara’s work led him through two post-docs, and then to a development job at Genentech. “I had to keep following the story—it almost obsessed me,” Ferrara says.

    Ferrara was hired to study the reproductive system, but Genentech allows researchers to spend up to 25 percent of their time on pet projects, so he continued to pursue angiogenesis. By 1989, he had isolated a pituitary-gland protein that stimulated vascular-endothelial cell growth. This protein, vascular-endothelial growth factor (VEGF), turned out to be a key angiogenic regulator. In 1993, Ferrara demonstrated that an antibody directed at VEGF could suppress angiogenesis and slow tumor growth in preclinical models.

    Avastin was approved in February 2004 for use with chemotherapy as a treatment for patients with first-line metastatic colon or rectum cancer—but that is far from the limit of its utility. Paradoxically, by focusing so narrowly on a particular target, Avastin became a drug with much wider potential use. VEGF plays a key role in turning on and off numerous cancers, and Genentech is pursuing a late-stage clinical-development program with Avastin to evaluate its potential use in adjuvant colorectal, renal cell (kidney), breast, and non-small-cell lung cancers. Avastin is also being evaluated as a potential therapy in prostate, ovarian, melanoma, and several types of solid-tumor cancers and hematologic malignancies.

    If the drug works as it should, it will become a compelling example of a new strategy for building blockbusters—what some people are calling the platform drug. Where the traditional blockbuster is built on a single application in a single clinical area, backed by the muscle of a Big Pharma sales force, the platform drug involves a particular mechanism that allows it to build sales volume by expanding into multiple indications, often in completely different clinical areas.

    “Avastin is probably the most powerful platform for cancer out there right now,” says Jan Malek, principal of PA Consulting. “Part of that reflects Genentech’s conscious commercial strategy for doing trials and getting registered for multiple indications. That’s very different than the single-indication blockbuster Wall Street has been enamored with—one drug, one sales force, one manufacturing process, one label,” Malek says.

    “Avastin’s broad spectrum of use was predicted and that prediction is now being borne out,” says Marc Tessier-Lavigne, senior vice president, research drug discovery at Genentech. For the first 12 months Avastin was on the market, it brought in $675.9 million, making it the most successful oncology product ever launched in the United States. But that’s a drop in the bucket compared to its potential: Wood Mackenzie says global annual sales for Avastin may reach $3.4 billion by 2009.

    A lot is riding on the success of Avastin—but it won’t come without its own set of challenges. First off, there is more competition in general in the oncology space. “We went from a handful of companies to now 30 or 40 companies in oncology,” says John Orwin, vice president, BioOncology sales and marketing. “There’s real competition for the physician’s time.”

    So far, Avastin has proved to be more effective at blocking VEGF than any attempts by competitors. But there have been a few close calls. “A lot of people are investing in oncology, but we’re also seeing a lot of drugs that seemed to be quite promising in Phase II, and in Phase III, something goes wrong,” Clark says. In particular, Novartis’ and Schering AG’s anti-VEGF, PTK787, failed to meet its primary endpoint of progression-free survival in patients with colorectal cancer, ending speculation that the drug would present a real challenge to Avastin, at least in the short-term.

    Next, Genentech must cross marketing hurdles, like  creating a base understanding of the drug and building the story for Avastin. More important, Genentech must develop a pricing strategy that keeps it competitive across a host of indications.

    Finally, reimbursement issues dog high-priced biologics like Avastin—even though the drug’s efficacy makes it a likely candidate for reimbursement from public and private payers. In Europe, for example, the price has deterred usage, despite the drug’s approval. “Virtually all patients in the US will be receiving Avastin as their first-line metastatic treatment,” says Joanne Graham, an analyst for Decision Resources. “And in the UK, virtually none are.”

Diversity of Approaches

Genentech’s portfolio of cancer drugs includes diverse mechanisms. Avastin is an antiangiogenic—designed to inhibit VEGF, a protein that plays an important role in maintaining existing tumor vessels—but other products invoke the body’s natural defenses against tumors or disrupt signaling pathways in various ways:

    Rituxan The first therapeutic antibody for use in cancer, Rituxan was initially approved in 1997 to treat certain forms of non-Hodgkin’s lymphoma (NHL). Rituxan works by recognizing and binding to a protein (called the CD20 antigen) on the surface of specific white-blood cells called B-cells, which are implicated in most forms of NHL. From there, it recruits the body’s natural defenses to attack and kill the marked B-cells.

    Herceptin Approved in 1998, Herceptin, has become a paradigm-shifting drug in the treatment of breast cancer. Used with patients with metastatic breast cancer whose tumors overexpress the HER2 protein, the drug has been a striking proof of the value of drugs that target diseases at the molecular level—and of the power of drugs paired with molecular-level diagnostics.

    Tarceva First approved in November 2004, Tarceva targets a gene that encodes the epidermal-growth-factor receptor (EGFR), part of the HER family, and which may block tumor cell growth. Tarceva is approved for the treatment of patients with locally advanced or metastatic non-small-cell lung cancer after failure of at least one prior chemotherapy regimen. FDA’s Oncologic Drug Advisory Committee recently recommended approval for first-line use, and for pancreatic cancer. FDA approval for that indication is expected early next month.

    Because these drugs use different mechanisms, they can potentially be combined, much like AIDS cocktails. Genentech is already exploring this. (See “Now the Evolution” section, page 80.)

    But several drugs, including Rituxan, are promising platforms for treating a full suite of immune diseases. A key insight initially came from the field: An investigator in London gave Rituxan to treat rheumatoid arthritis (RA) in five patients who had failed all available therapies. They had a dramatic positive response. On August 31, 2005, Genentech and partner Biogen Idec submitted a Biologics License Application (BLA) to FDA for Rituxan’s use in treating RA.

    “Rituxan is the poster child here, because not only have we taken it into clinical trials for RA, but also other autoimmune disorders like lupus and multiple sclerosis,” Tessier-Lavigne says. “Immune disorders that were thought to be mediated by T-cells or other cells were shown to be ameliorated by hitting B-cells with Rituxan, leading to a re-evaluation of the preclinical biology. People realized, lo and behold, there are conversations between B-cells and T-cells that we didn’t anticipate. This was a case of the clinic informing the basic science, but my guess is this will be the rule rather than the exception.”

Stretch Development Dollars

In 2004, the company spent $947.5 million on R&D, ranking 14th in the industry, even among Big Pharma. When that analysis was weighted to look at research dollars in relation to sales, Genentech came in second. Finding new indications for existing drugs helps Genentech get more out of that spending. Indeed, Genentech’s growth over the next few years will come not from new drugs, but rather current products. Avastin and Rituxan are forecast to generate $2 billion each by 2007 (some forecasts skew even higher). Together, those two products will account for 59 percent of total sales by 2010, according to Datamonitor.

    Smart development “Genentech already knows the drugs are safe and tolerable, so they can avoid doing a Phase I and possibly even a Phase II study,” says Jason McKinnie, a research analyst for Frost & Sullivan. “This gives them the ability to move quickly into other cancers, because they don’t have to spend as much R&D time on a drug. Avastin looks like a brand-new drug for each cancer it’s in, but the development time and expense are cut way down. They are now looking to do that and get into different cancers with Tarceva.”

    Here, the biggest challenge is which indication to go for first: “On the face of it, you would want to go for the biggest potential single indication market,” says PA Consulting’s Malek. “But Genentech has been smart, because the first indication FDA approves establishes the label and the side-effect profile of the drug. You carry a lot of that with you to other indications.”

    Sales-force efficiencies Platform drugs  in oncology exploit synergies between tumor types. From a commercial point of view, these tumor types may seem like new markets, Clark says, but it allows a drug with several indications to be sold by the same representatives, potentially offering a major saving.

    “A lot of oncologists are community oncologists that have one patient with breast cancer, the next with lung cancer,” Clark says. “It’s the same customer—we don’t really have to hire new reps.”

    Smaller clinical trials Diagnostics—the telltale biomarkers that help determine how a drug actually works and which patients are the most appropriate—pave Genentech’s path toward targeted therapies, and each development program is now required to have one.

    “We have documented that if we hadn’t had the HER2 diagnostic, it would have been very difficult to show that Herceptin works,” Tessier-Lavigne says.     “If you set up a clinical trial where only a quarter of the patients are responding and the effect is washed out by the three-quarters who aren’t responding, then the signal gets lost in the noise. So the diagnostic really was required for the success of the drug.”

    Clinical trials change when you know the disease is being driven by one pathway. “Theoretically, then clinical trials can get much smaller because the treatment effect is going to be really big and therefore, much easier to see,” says Hal Barron, MD, senior vice president of development and chief medical officer at Genentech.

Translational Medicine   

While Genentech uncovered a few lucrative platform drugs, it has carefully crafted drugs that remain true to the idea of targeted medicines—a legacy that began with Herceptin’s initial approval in 1998. While the breast-cancer market is big—close to 200,000 diagnosed cases in 2004, according to Decision Resources—Herceptin is only indicated for about 25 percent of breast-cancer patients, those who overexpress the HER2 gene.

    “The mercenary view is you are dividing the market into two or into five, and therefore your drugs just got half or a fifth of the size,” Clark says. “In some ways, that’s true. But we think you should be highly effective in that subset, and therefore, it should be a very successful drug.”

    Herceptin demonstrates that a narrowly targeted drug does not have to be a small drug. Recent Phase III trials showed that when Herceptin was given to women with early-stage HER2-positive breast cancer in the adjuvant setting following surgery, the risk of recurrence was reduced by 52 percent compared to chemotherapy alone. That’s good news for patients, but also for the company: It means that a relatively small market can be expanded by treating earlier-stage patients, and treating them longer compared with the late-stage metastatic market. Datamonitor projects the brand’s sales to nearly triple from $795 million in 2004 to an estimated $2 billion by 2009.

    “They have had the confidence to go for niche markets,” says Decision Resources’ Graham. “That’s a lesson to the major pharma companies—you don’t have to have every patient treated to be successful.”

    Creating highly targeted therapies has bathed Genentech in the warm glow of virtue: They are on the right side of the evolving healthcare landscape. When researchers presented the adjuvant data on Herceptin at the 2005 American Society of Clinical Oncology (ASCO), ten thousand people attended the presentation. Although investigators missed ASCO’s deadline for submission of abstracts, the society realized the data’s importance and added a stand-alone symposium.

    “The moderator said, ‘Science has spoken and we should go forward and change the way we treat breast cancer,’” Clark says.

    Ultimately, this sense that a product grows inevitably from rigorous science will have a major impact at the commercial level. Tailoring drugs to patients’ genotypes offers rational therapy with fewer side effects. And public and private payers are likely to reimburse for the drugs because Genentech has proven they work in rigorous clinical trials. (See “Robust Clinical Trials,” page 80.) With such convincing results, Graham says, patients can argue that it’s unethical for payers to withhold treatment.

    “Getting an aggregate benefit isn’t acceptable,” Barron says. “We’ve got to focus our biologics because they’re so expensive to make. We need to ensure that patients are getting the best benefit/risk [ratio] they can, and therefore, cost-effectiveness of that therapy becomes much more acceptable.”

    Marketing to subpopulations Genentech was able to establish a link between Herceptin and commercial success through diagnostic testing. “We created a market for HER2-positive breast cancer in the sense that the diagnostic allows us to identify which patients will benefit from Herceptin,” BioOncology’s Orwin says.

    That strategy may not be so easy to apply to other targeted therapies, particularly Tarceva. Although Genzyme is planning on commercializing a diagnostic kit to detect EGFR gene mutations later this year, it won’t define the exact patient population. Recent studies show that Tarceva works in both EGFR-positive and EGFR-negative patients.

    “Some patients get a huge amount of benefit from Tarceva,” says Gwen Fyfe, MD, vice president of clinical hematology and oncology. “Then there’s a group that gets some benefit. And then there are those who get no benefit. The problem is that while we have some indication of who gets the most benefit—because their tumors shrink—we don’t have any markers for those patients in the middle who get a modest-to-moderate benefit.”

A look at competitor Iressa (gefitinib) shows the importance of better understanding the patient profile: Although a study showed that AstraZeneca’s drug did not significantly prolong survival, it clearly works in some patients. FDA told physicians to continue prescribing the therapy in patients who have shown improvement, but to avoid dispensing it to new patients.

Now the Evolution

Genentech has already created a revolution with innovative drugs. Now, its products and new indications will define the evolution as the company applies its research base.

    Take Lucentis, which Genentech plans to submit to FDA this December for treatment of “wet” age-related macular degeneration (AMD). Genentech began investigating this disease because of its association with VEGF.

    “You don’t just say, ‘Boy, AMD’s a big disease. Let’s throw something at it,’” Barron says. “Rather, you follow the science.”

    The company found a proliferation of blood vessels in the macula part of the retina in people with macular degeneration. “We said, ‘Let’s start exploring,’” Barron says. “Now, with AMD, you wouldn’t confuse it with a tumor, but it’s like all the vasculature you see in the tumor. And if you sample the vitreous, VEGF levels are very high. So in this case, the VEGF level was a nice way of identifying a population.”

    And a nice way of leveraging the science. Lucentis is a modified antibody fragment, designed for local delivery into the eye. According to Deutsche Bank biotech analyst Jennifer Chao, Lucentis will bring in $150 million in sales by 2007, and $310 million by 2008. Even though Pfizer and Eyetech’s AMD drug Macugen (pegaptanib), another VEGF inhibitor, beat Lucentis to the market, Genentech’s drug looks like it will work better than Macugen: It was shown to stabilize or improve vision in 95 percent of patients, and its benefits were reported widely as superior to Macugen’s.

    Avastin has additional mileage. “We have a number of pre-clinical programs that try to add onto the effects of Avastin and hit factors that collaborate with VEGF in tumor angiogenesis—you know, give them the one-two punch,” Tessier-Lavigne says.

    “Their dream, I’m sure, is cocktails of therapies given at one time—all made by them,” Graham says.

    Currently, the company is testing Avastin in combination with Tarceva for several indications. That may increase Tarceva’s effectiveness, especially as it faces competition from a new class of next-generation EGFR inhibitors that may cause longer-term submission of lung cancer. But McKinnie notes that there are additional obstacles to the brand’s success.

    “There is a strict patient population that is allowed to take Avastin for lung cancer,” McKinnie says. “They can’t have brain metastasis, and they must have a certain type of lung cancer, even in the non-small-cell setting. Because of that, there may not be as much growth in the lung-cancer field as everyone says.”

    Another combination that seems promising—at least in theory—is the use of Avastin with a therapy that targets EG-VEGF (endocrine gland vascular endothelial growth factor), a protein Ferrara identified in 2001 through SPDI (Secreted Protein Discovery Initiative), the company’s systematic approach to identifying molecules thought to be important in regulating disease. EG-VEGF, as it turns out, is responsible in forming new blood vessels in endocrine-specific tissues.

    “Take the example of the corpus luteum on the ovary,” Ferrara says. “We propose that VEGF is very much a primordial molecule—it is probably essential for development of the blood vessels and the maturation of the vasculature. And then later on, EG-VEGF is expressed. Perhaps if you combine anti-VEGF with an anti-EG-VEGF, it could lead to better treatment and more complete inhibition—this is a hypothesis that we are working on right now.”

    Long-term growth Today, Genentech sees few threats to its patents. Several sources say the company won’t face patent issues on any of its products in the next five to 10 years at least.

    But development times are long, and the possibility of FDA establishing a biogeneric pathway means Genentech must prepare today for the company’s long-term growth—what senior management sees as the principal challenge for the organization. To do that, the company aims to diversify its reliance on sales from currently marketed products to the next round of high-tech therapies.

    In oncology, Genentech has a few promising early-stage candidates, including Apo2L/TRAIL, which represents a new approach to treating cancer through programmed cell death, and Omnitarg (pertuzumab), the first in a new class of agents known as HER dimerization inhibitors (HDIs).

    “The oncology world will only be so big,” Clark says. “Therefore, we must demonstrate that we have a scientific ability and a development organization that can supply products that we need to be able to grow in 2010 or 2015.”

    For its future, the company is looking toward the field of immunology. Genentech already established its foothold in 2003, when it launched Xolair (omalizumab), a humanized antibody designed to target IgE, a key underlying cause of the symptoms of allergy-related asthma, and Raptiva (efalizumab), a humanized antibody that blocks the T-cells that lead to the development of psoriasis. Although those drugs face strong competition, the approval of Rituxan to treat several immune conditions, and development of two pipeline products—BR3-Fc and Anti-NGF—should help bolster that franchise.

Execution Game

“We remain a decidedly science-based drug-discovery organization, going from a deep understanding of the basic biological mechanisms, and turning those insights into medicines,” Tessier-Lavigne says. “That’s what’s possible today, and we’re only limited by our imagination and our ability to organize and execute. And we’re working hard to make sure that we do that as well as possible.”

    Indeed, many analysts see Genentech as the darling of the industry. According to Standard & Poor’s DiLorenzo, executing product filings and launches in quick succession is key. For Genentech, that is easier said than done. The company is addressing the challenges that it will face in recruiting people, establishing processes, and building up its commercial know-how to handle the execution game.

    Recruitment Given the company’s location on the West Coast—across the country from pharma’s heartland—recruiting has always been a challenge. That becomes an even bigger issue as the company enters this next growth phase and builds out sales forces for Lucentis, and other products in new areas.

    Genentech hired 1,500 people in 2004, a 23 percent increase over 2003, yet managed to create diversity among top management: At year-end 2004, 44 percent of officers and managers were women, and 29 percent were minorities. Can the company keep the science-focused culture alive as it grows by thousands of people in the coming years?

Processes Already, Genentech’s manufacturing plants need to run full-time to meet demand. With that demand expected to soar, the company had to take steps to ensure its future supply. In June, it bought a manufacturing plant from Biogen Idec, in Oceanside, CA, and began expanding its already existing Vacaville, CA, manufacturing plant. When that expansion is completed, the Vacaville site will be the largest biotechnology manufacturing facility in the world. By early 2007, Genentech expects to have 370,000 liters of licensed manufacturing capacity, which will represent about 35 percent of the world’s capacity for biologics manufacturing.

Commercial know-how Genentech must leverage the marketing platforms surrounding drugs that have broad utility, like Avastin, with plans for drugs with narrowly defined indications, like Herceptin—and use them in combination.

    “As our understanding of the molecular basis of the disease increases and our therapies become more specific and more tailored almost by definition, they begin to define smaller markets,” Orwin says. “It’s somewhat inevitable that the therapeutics in cancer are going to become more selective over time. That’s not to say one wouldn’t look for a blockbuster if there was one out there. But it means that companies have to be able to handle a blockbuster opportunity like Avastin, and to develop and commercialize highly-targeted therapies that might only benefit a subpopulation of patients.”

    On the Genentech campus, there’s a life-sized bronze sculpture of two men at a table, beer glasses in their hands, one with long, curly hair and an academic air, the other leaning forward in a shirt and tie, eager and intent. The piece commemorates the birth of the company in 1976. Herbert Boyer, the academic, was a professor at UCSF and the director of the graduate program in genetics. Working with Stanley Cohen, he inserted frog DNA into a bacterium, creating the world’s first recombinant-DNA organism. A 29-year-old venture capitalist named Robert Swanson, convinced that this type of technology could form the basis of a company, took Boyer out for a beer, and three hours later Genentech was born. The sculpture captures not just a moment, but an ideal—the scientist and the businessman, distinct, engaged, excited, and in harmony. It’s a hard idea for a growing company to continue to embody, but so far at least, Genentech is on track.