Where ImClone Went Wrong

What biotech--and Big Pharma--could learn from its debacle

How ever did it come to this? Since the Food & Drug Administration refused to accept an approval application for ImClone's promising new Erbitux cancer drug on Dec. 28, the company has been in something close to a free fall. Once one of the hottest biotech stars, ImClone Systems Inc. (IMCL ) is grappling with a congressional investigation, Securities & Exchange Commission and Justice Dept. inquiries, and more than a dozen shareholder lawsuits. And now, ImClone development partner Bristol-Myers Squibb Co. (BMY ) is demanding that the company's founding brothers, CEO Samuel D. Waksal and COO Dr. Harlan W. Waksal, step aside. If they refuse, the pharma giant says it will walk away from its $2 billion commitment to ImClone.

But Bristol-Myers CEO Peter R. Dolan has some explaining of his own to do, and so may the FDA. Sources close to the clinical trial say Bristol surely reviewed all communications between ImClone and the FDA, so it should have been aware of any problems. Yet both Bristol-Myers and ImClone insist they were stunned by the FDA's action. How that could be is one of the deep mysteries of this mess.

Clinical trials are often designed in close collaboration with the FDA, and the agency documents every conversation it has with company executives. Hundreds of pages of such memos have been filed with the congressional committee investigating ImClone. Erbitux' trial was in progress before ImClone approached the FDA, but the agency still had to review and sign off on the design, which tested Erbitux on 120 very sick colon cancer patients in combination with standard chemotherapy. That's why congressional investigators are now asking how Imclone's application could have been so out of sync with the FDA's requirements.

ImClone might take cold comfort in the fact that it is not the first biotech company to run into trouble on the FDA's doorstep. Over and over, drug approvals are delayed because the agency wants more data. In the past several months, Corixa (CRXA ), Chiron (CHIR ), Northfield Laboratories (NFLD ), Genzyme (GENZ ), and Genentech (DNA ) have all had setbacks at the FDA with drugs trumpeted as promising. Several other companies have announced delays in filing their FDA applications while they tinker with late-stage clinical trials.

All of these companies must constantly wrestle with reconciling the hyperdrive demands of a cash-hungry business and medical research that is slow and rarely definitive. It can take decades--and hundreds of millions of dollars--to develop a new drug. Many biotech companies, ImClone included, pin all their hopes on just one compound and are regularly forced to go hat in hand to venture capitalists and the stock market to keep that drug in the running. In the background are desperately ill patients clamoring for effective treatments, adding to the urgency.

"I think a lot of biotech companies get into survival mode instead of success mode," says Richard B. Brewer, CEO of Scios Inc. (SCIO ) in Sunnyvale, Calif. Scios' Natrecor heart failure drug was delayed for a year and a half after the FDA unexpectedly asked for more trials. It finally won approval last August. "They forget that the FDA is not here to get your drug approved. It's here to protect patients."

ImClone seems to have made all the classic mistakes of a biotech company hungry for a quick drug approval. It rushed through its FDA application, kept clinical trials as small as reasonably possible to save money, and hired board members renowned for their scientific credentials--meant to impress investors--rather than their experience with the FDA. Only two of ImClone's directors came from the drug industry--former Ciba-Geigy U.S. Chairman Richard Barth and and former Bristol-Myers Vice-Chairman William R. Miller. ImClone's chairman, Robert F. Goldhammer, is a partner at money-management firm Concord International Partners.

ImClone may have had another liability: its brash and boastful CEO. Sam Waksal, 53, has long raised eyebrows on Wall Street with his activities outside ImClone. He hangs out with celebrities, throws high-profile parties in his Soho loft, and has made poor investments in restaurants. He and his brother borrowed more than $33 million from ImClone in July 12 to exercise options and warrants on 2.5 million shares. He has also been a relentless promoter of Erbitux. Throughout last year, he assured investors that the drug would win speedy FDA approval and be available by mid-2002.

It's easy to understand his hurry: ImClone has not turned a profit in its 17 years of existence and has spent close to $200 million and nine years developing Erbitux. His enthusiasm is also understandable. Cancer specialists have been buzzing about Erbitux since 1999, when the drug, combined with chemotherapy, showed notable efficacy against late-stage colon cancer. Granted, it isn't even close to a cure. But tumors shrank in 22.5% of the participants in a 120-patient clinical trial, a high response rate for a cancer drug. Plus, to date, Erbitux has produced no severely debilitating side effects.

The FDA was impressed enough by the early data that it granted Erbitux fast-track status in February, 2001, meaning it would review the drug within six months after an application for approval was filed. ImClone executives promised investors that the application would be submitted by July, even though data from the 120-patient trial was only publicly released in May. That was an ambitious plan, since FDA applications are notoriously difficult to put together, typically requiring a truckload of verifying data.

Then came the first hint that the approval process would not be a smooth one. ImClone started filing its application on June 28. But it was a so-called rolling application, meaning sections would be submitted as they were completed. Ultimately, ImClone did not complete its application until Nov. 1, setting off the FDA's own 45-day clock to decide if the application was adequate.

Meanwhile, ImClone had already won some validation on the business side. Sam Waksal announced in September that Bristol-Myers Squibb would buy a 20% stake in the company for $1 billion. Shareholders got a windfall, since Bristol was paying $70 a share, a $20 premium over the trading price, just to get its hands on 40% of the marketing rights to Erbitux. Waksal and his brother Harlan together made $111.3 million when they tendered 20% of their shares. Plus, Bristol promised $1 billion more in payments to be parceled out every time ImClone met certain key milestones in the drug's development. One of these payments--$300 million--would fall into ImClone's coffers as soon as the approval application was accepted by the FDA. Since the FDA rejects only a handful of such applications each year, ImClone likely felt the money was a lock.

But the FDA's 45-day deadline, Dec. 15, came and went with an ominous silence. When the agency announced on Dec. 28 that it would not accept ImClone's application, the boom fell--sort of. Sam Waksal insisted that the FDA's main concern was that the company did not include enough data on the disease status of the patients in the trial and that ImClone could easily provide that data. It was another clue that the application was rushed: Industry analysts say it is unusual not to include information on the status of each patient in a new-drug application.

ImClone still might have recovered its footing with the investment community--if the highly secret FDA letter spelling out the agency's objections hadn't been leaked. As reported in The Cancer Letter, an industry newsletter, the FDA said the clinical trial used to support the application was not adequate and that a new trial could be needed before the drug could be approved. Even more damning, it said the company had been told repeatedly about problems with its clinical trial, even though Waksal had said--and continues to say--that the company had no idea the trial would be deemed inadequate.

Is he telling the truth? At least 12 shareholder lawsuits charge that he isn't. The answer may be clearer once the company meets with the FDA in late February to determine how best to go forward. Right now, the FDA won't comment. Richard C. Mulligan, a genetics professor at Harvard Medical School and a member of ImClone's scientific advisory board, defends Waksal: "I find it inconceivable that there was any underlying willful deception." Instead, says Mulligan, who was not involved in developing or testing Erbitux, "it is quite likely that there were errors made in preparing the application."

Biotech executives say such errors are common, in part because FDA communications can be confusing to a novice. Scios CEO Brewer notes that the FDA can easily change gears midstream--as it did with his company's Natrecor. After the drug was recommended for approval by the agency's outside advisory committee, the FDA's own reviewers rejected it and asked for costly new clinical trials that would show the drug's efficacy in head-to-head comparison with an existing treatment--rarely done for drugs pre-approval. "You can say that's not fair or that's not what you told us in the first place, but they don't care about being fair," Brewer says. "They'll just say, `things change'--and they're right."

Doctors involved in the Erbitux trial who were canvassed by BusinessWeek all say they remain enthusiastic about the drug's potential and are disappointed by the way things have changed. "There is no doubt that the compound works," says Dr. Robert J. Mayer, director of gastrointestinal oncology at Dana-Farber Cancer Institute in Boston, who has no financial stake in ImClone. "We have had people who benefited quite dramatically."

Mayer was an investigator in a trial of 57 patients who received Erbitux without chemotherapy, the results of which were also submitted to the FDA. About 11% of those patients responded, lower than the 22.5% response rate when Erbitux was combined with chemo but still considered good for a cancer drug in late-stage cancer patients. "The greatest frustration that people like me have is that there is a highly useful drug gathering dust in a back room due to the regulatory morass," says Mayer.

Bristol thinks it can pull Erbitux out of this morass, particularly if the Waksals get out of the way. Doctors investigating the drug say it is still the most promising compound around for colon cancer, a particularly deadly disease once it spreads. If more clinical trials are needed, there are plenty of patients eager to participate, they say, and trials could be completed by the end of the year.

Or, Bristol could base its FDA application on trials being conducted by the German company Merck, which holds the European rights to Erbitux. Merck is testing the drug against both head and neck cancer and colon cancer, and "the studies look good," says a spokesperson. It expects to present its results to the European Union by July or August. So biotech analysts are still confident that the drug will win FDA approval by mid-2003. Although Astra-Zeneca PLC has a similar drug, Iressa, in clinical trials that may arrive on the market first, that compound is being tested for lung cancer and may not be appropriate for colon cancer, doctors say.

Meanwhile, patients suffer. Dana-Farber's Mayer complains that "yes, the stock is down. Yes, there are unhappy people on multiple levels. But my concern is that this is an effective drug--and it is not getting to patients." He can only hope that Bristol-Myers does a better job the second time around.

By Catherine Arnst in New York, with John Carey in Washington and Jack Ewing in Frankfurt

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