ImmunoGen’s Mitch Sayare on What Sustains a Company—and its CEO—Through Decades of Starting Up
Call it the 27-year-old startup: ImmunoGen, one of biotech’s oldest companies, has yet to get a drug on the market more than a quarter century after its formation in 1981. This year may mark a long-awaited turning point for the Cambridge, MA-based company, though. If all goes well, compelling efficacy data will emerge from ongoing Phase 2 trials with ImmunoGen’s cancer drugs, paving the way for pivotal trials that could lead to the FDA-approved promised land. When I dropped by ImmunoGen (NASDAQ: IMGN) last week, CEO Mitch Sayare began by acknowledging that “I’m more optimistic than I’ve ever been” since taking the company’s helm in 1986. But he couldn’t resist adding a verbal version of knocking on wood: “My optimism is tempered by experience.”
He might have said bitter experience: Few cancer therapies have looked as good in theory—and have proved as frustrating in practice—as ImmunoGen’s. Spun out of Boston’s Dana-Farber Cancer Institute, the company has pioneered “immunoconjugates”—two-part medicines consisting of monoclonal antibodies, the guided missiles of biology, chemically attached to highly toxic molecules designed to amplify tumor-killing power. Such drugs can target cancer cells with poisons that are far more potent than traditional chemotherapeutics, in principle enabling tiny doses with relatively minor side effects to knock out tumors.
Unfortunately, for years the approach worked just well enough to lure ImmunoGen down a tortuous dead end. The company spent most of its early life trying to demonstrate efficacy with conjugates employing a form of ricin, a toxin derived from castor beans that’s said to be 12,000 times as deadly as rattlesnake venom. Monoclonals studded with ricin-derived molecules duly “killed targeted cells in patients,” says Sayare. “But what defeated us was ricin’s immunogenicity”—after one or two doses, patients developed immune responses to the ricin-ferrying drugs that cleared them from the blood before they could do any good. “You can’t treat cancer with an agent that you can use only once,” he adds, “because even if you can kill 99 percent-plus of cancer cells with the first dose, there will still be billions left.” ImmunoGen shelved its ricin conjugates in the mid-1990s after more than a decade of work on them.
Such key-project failures have been known to kill small biotechs. Indeed, “there were times in the late ’90s when we came very close to the big pit,” recalls Sayare. “We looked down and were able to walk away.” Pulling ImmunoGen back from the brink was a chemical even more marvelously toxic than ricin: maytansine. Isolated in 1972 from an Ethiopian shrub, maytansine showed anti-cancer promise in 1970s animal studies, prompting the National Cancer Institute to spend almost a decade testing it in patients with various kinds of tumors. The substance by itself, however, proved to have more toxic side effects in people than in animals, and the NCI put it aside in the mid-1980s.
A few years later, ImmunoGen scientists, casting about for alternatives to ricin, were intrigued to learn that the NCI had ranked maytansine as the strongest cell killer ever tested against tumors. Their interest grew when a gifted chemist at the company, Ravi Chari, noticed that maytansine’s structure made it amenable for use in conjugates. By the time ImmunoGen gave up on ricin, it had begun developing monoclonal antibodies tagged with maytansine derivatives. The drugs quickly showed substantial preclinical promise, putting ImmunoGen in a position to cut sorely needed new deals with pharma partners.
Sayare faced a hard sell after the ricin crop failure, though. He recalls spending over two years trying to convince Genentech to try a “maytansinoid” conjugate based on Herceptin, the San Francisco biotech’s groundbreaking monoclonal antibody for breast cancer—test-tube studies at ImmunoGen had suggested such a drug would work well. In 1999, Genentech finally asked ImmunoGen to stir up a vial of Herceptin-based conjugate for preclinical tests. “After we sent it to them, they went radio silent on us for six months,” says Sayare. “I couldn’t even get my calls returned. Then suddenly they were all over us, wanting to do a deal.” It was signed in 2000, and since then Genentech has advanced a Herceptin-based conjugate dubbed T-DM1 into a closely-watched Phase 2 trial following a Phase 1 that yielded perhaps the most promising data to date from any conjugate: Twelve of 15 advanced breast-cancer patients who got the “maximum tolerated dose” had a partial response (tumor shrinkage) or stable disease (tumors neither increasing or decreasing) for up to about nine months—a markedly better result than Herceptin alone has achieved in such patients. Importantly, the patients’ tumors had already failed to respond to up to half a dozen other regimens, including Herceptin itself.
Genentech also has disclosed that it is investigating over 20 conjugates, including several from ImmunoGen, giving such medicines Next-Big-Thing status in oncology. Just one conjugate has reached the market to date: Wyeth’s Mylotarg, approved in 2000 to treat acute myeloid leukemia, a bone-marrow cancer.
But ImmunoGen’s new conjugates have also traveled a rocky road. In June 2002, GlaxoSmithKline pulled the plug on one that it was developing with the Cambridge firm, and almost simultaneously Genentech delayed a planned trial with T-DM1 in order to complete extra preclinical tests. The double whammy nearly chopped ImmunoGen’s market cap in half. But a year later it bounced back, forming a major collaboration with France’s Aventis, now sanofi-aventis, that so far has produced three drugs in clinical trials. All told, five of ImmunoGen’s conjugates have reached the clinic, including two it wholly owns; one other ImmunoGen drug, licensed by sanofi, is a nonconjugated “naked antibody” in early-stage clinical testing against various cancers.
The research portion of the sanofi collaboration winds down this summer, but by next year, says Sayare, ImmunoGen hopes to form at least one new Big Pharma deal and to announce that three to five more of its drugs have reached the clinical-testing stage. Meanwhile, further clinical results are expected to be reported this year on T-DM1, and on two wholly-owned ImmunoGen conjugates in tests against multiple myeloma, a blood-cell cancer, as well as gastrointestinal and other cancers. The emerging data may well lift ImmunoGen’s stock price, which has been languishing under $4 a share, paving the way for a $75 million stock offering—the company filed shelf registration for the offering last July. Sayare declines to guess when T-DM1 or another ImmunoGen drug might get FDA approval. But he notes that if one of the ongoing Phase 2 trials with its medicines yields sufficiently compelling data over the next year or so, it should be possible to convert the trial into a pivotal study, considerably shortening the march to market.
With light glimmering at tunnel’s end, it seems a good time to ask Sayare how ImmunoGen has managed to keep chugging so long. “I knew you’d ask that, so I’ve been thinking about it,” he says. “In biotech companies, I think there are three things that sustain people. Number one is the altruistic part. There isn’t anybody here who hasn’t been affected by cancer in their families or their own lives.” (Sayare’s mother died “quite young” from breast cancer.) “Number two is ego. You can’t get away from it. I mean, who wants to walk away from a disaster? And number three is economic. I can’t deny it, there’s money to be made here for all our stakeholders.”
As it happens, four years after joining ImmunoGen Sayare also took up a hobby that’s helped him cope: amateur astronomy. Gazing out a few million light-years with his six-inch reflecting telescope shrinks earthly matters down to size, he says, “and for one night I can get a fabulous escape from all the problems at ImmunoGen.” Point taken: Maybe the shopworn motto “to the stars through difficulties” should be renovated to read: “To the stars because of difficulties.”