Cancer Man

In a dimly lit conference room in the basement of Orlando, Fla.’s swank Peabody hotel, Patrick Soon-Shiong, the richest man in the drug business, is working his magic on a dozen doctors who are in town for the year’s biggest medical conference.

Over a breakfast buffet, Soon-Shiong, 57, urges the oncologists to enroll their patients in a clinical trial of his cancer drug Abraxane, to test it against pancreatic tumors, which have resisted every new drug for decades. It’s a tough argument. The common wisdom about Abraxane, made by Abraxis Bioscience of Los Angeles, is that it’s just a souped-up version of another drug.

But Soon-Shiong soon takes another tack, saying he is going to share an idea that has been “whirling in my head for two years.” He asks the doctors to ponder one of the biggest puzzles in biology: Why is it that some cancer cells escape tumors to take up residence elsewhere in the body, allowing the disease to spread? This process, metastasis, is part of what makes cancers deadly.

Soon-Shiong argues that his decades of work on Abraxane provide part of the explanation for metastasis. And it just so happens that Abraxane is an ideal weapon against cells that have turned on their metastatic machinery. By the end of his talk the doctors are energized, if not totally convinced.

“The hypothesis is a great hypothesis,” says Ben Ebrahimi of Wilshire Oncology in La Verne, Calif., one of the doctors who was present at Soon-Shiong’s talk. “We need the clinical data to back it up.” As always, Soon-Shiong remains undaunted. “If this is real,” he says, “it is profound.”

Soon-Shiong is prone to exuberance. When Abraxane was approved in 2005, he hailed it (and still does) as a breakthrough. But it generates so-so sales of $300 million annually, roughly a tenth of what cancer drugs like Avastin and Taxotere generate–and far less than either Soon-Shiong or Wall Street projected a few years ago. In 2008 Abraxis Bioscience, the company he founded and of which he owns 80%, lost $278 million, including $159 million to dissolve a marketing partnership with AstraZeneca ( AZN – news – people ), on revenue of $345 million and is projected to lose $90 million more this year. Investors have driven down its stock by 60% this year. “I think he’s viewed as ‘What’s best for Patrick is what carries the day,’” sniffs Michael King of Merriman Curhan Ford, the only analyst covering the company.

Read All Comments

No matter. Soon-Shiong, worth an estimated $4 billion, has worked an age-old formula for getting rich: Don’t give up equity. He has maintained ownership of the drug he invented; until he sold it last year he gave up little equity in the hospital generic injectables business he founded in 1997. He’s the only drug researcher who’s a billionaire. The other members of The Forbes 400 to come from the drug business are an expert marketer and executive (Michael Jaharis), a venture capitalist (Randall Kirk) and the founder of a generics firm (Phillip Frost.)

Soon-Shiong was born in South Africa in 1952 to parents who had fled China during World War II. At 23 he got an M.D. from the University of the Witwatersrand in Johannesburg. After a residency in surgery, he was recruited by UCLA Medical School. He soon displayed a zealousness for finding cures for illnesses that other doctors prefer to treat. He performed the school’s first pancreas transplant for a woman with diabetes, a radical–and successful–surgery for something usually treated with insulin.

Impatient with the pace of diabetes research, he left UCLA to start his first company. He performed a pioneering implant of insulin-producing cells in 1993. He gained widespread notice, but some colleagues weren’t impressed and the procedure wasn’t replicated.

Soon-Shiong soon went in another direction. His diabetes work led him to understand how to create a drug that would get directly into cancer cells, which were a lot like the stem cells he was studying in diabetes. In 1993 he invented Abraxane. It’s a variation on the well-established breast cancer drug Taxol, which works by inhibiting cell division.

Rejecting money from venture capital funds, Soon-Shiong found a different way to pay for the expensive clinical trials required to get Abraxane approved by the Food & Drug Administration. At the time, he was starting a company to sell injectable cancer drugs and antibiotics to hospitals. He used profits from that company, American Pharmaceutical Partners, to fund Abraxane’s development in breast cancer. APP went public in 2001.

Soon-Shiong said he would get Abraxane approved via an FDA rule that would let him use just one clinical trial. Wall Street was skeptical about the shortcut. At one point in 2003 the short interest in American Pharmaceutical Partners climbed to 100% of the available shares. When the FDA approved the drug in January 2005, the stock jumped 30% in a day. A gloating Soon-Shiong gives away plastic mementos displaying that day’s stock chart.

He sold APP last year to Fresenius, a German dialysis company, for $5.6 billion, of which he got $3 billion. His 80% share in Abraxis Bioscience is worth another $860 million.

A 2005 article in the Annals of Oncology called Abraxane “old wine in a new bottle.” The old wine, paclitaxel, was cancer’s first $1 billion drug and sold by Bristol-Myers Squibb ( BMY – news – people ) under the Taxol name. Since the patent expired in 2000, the market has shifted to the generic version.

Taxol comes dissolved in a solvent so potent it melts standard iv tubing. Abraxane instead coats the drug molecules in albumin, one of the main proteins in egg white. This prevents any allergic reaction and allows the drug to be given in 30 minutes instead of three hours. Abraxane’s wholesale cost is $5,100 a dose every three weeks, which is 16 times as much as generic paclitaxel. “Definitive studies to show it is better than Taxol have yet to be done,” says Eric Winer, a top breast cancer researcher at the Dana-Farber Cancer Institute. Nor has Abraxis finished the studies required to get Abraxane approved for use in any other cancer. Results in lung cancer are expected soon.

But Soon-Shiong says his drug is more effective than Taxol, precisely because of a quirk in the albumin coat. Therein lies Soon-Shiong’s next scientific adventure: stopping cancer from spreading from, say, the breast, where it won’t kill, to the brain, where it will.

He argues that a chemical called Sparc (secreted protein acidic and rich in cysteine) is the culprit, leading smoldering cancers to suddenly spread like wildfire. Sparc has long been implicated by researchers in the spread of colon cancer to the liver and may be similarly predictive in melanoma, breast cancer and glioblastoma, a type of brain tumor.

Soon-Shiong argues that when a cell starts making Sparc, suggesting it’s about to spread, it also starts sucking in albumin-soaked Abraxane, which, he says, will kill the tumor. By studying whether Abraxane works in pancreatic cancer, where many cells are churning out Sparc, he will be able to help solve the mystery of metastasis. “We’ve been trying to shoot cancer from across the ocean,” he says. “I think we have penetrated all the way to the heart of the tumor itself.”

Patients whose cancers are churning out the Sparc protein usually have a worse prognosis, but with Abraxane they may do just as well, Soon-Shiong argues. Preliminary analysis in four different cancers indicate that Sparc-positive patients are as much as three times more likely to benefit.

Soon-Shiong offers up a list of scientists who he says are excited by his theory of metastasis. Wishful thinking. Several protest that the evidence needs to be published and that they have not seen data but just had interesting discussions. Magic bullets for understanding metastasis “have been proposed again and again, and nothing has ever held up,” says Richard Klausner, former head of the National Cancer Institute and now a principal at the Column Group, a venture capital firm. “It’s a great idea, but it’s not like when I heard the biology of it I’ve said, ‘Aha, this makes sense.’ Let’s see the data.”

Now Soon-Shiong is starting yet another company: Abraxis Health will get a $25 million equity infusion and a $200 million line of credit from Abraxis Bioscience to delve into the diagnostics business, probably beginning with a test for his metastasis marker. “If the Sparc hypothesis proves to be true and it changes the course of medicine, it would be more satisfying than all the money we ever made,” he says.

Soon-Shiong, who is married to Michele Chan, a 1980s television actress (in shows like MacGyver), says he plans to start giving more to charity now that he has sold APP. The first big gift, of $100 million, went to Saint John’s Health Center, which is associated with UCLA.

He has given up the chief executive spot of Abraxis Bioscience to Lonnie O. Moulder, who built tiny MGI Pharma, maker of an antinausea drug, into a successful biotech. In September Abraxis Bioscience picked up analyst coverage for the first time in a decade. King, at Merriman Curhan Ford, gave the company a buy rating–predicated on the beaten-down share price and the idea that Soon-Shiong will be absent from the company he founded. Soon-Shiong shrugs and says, “I think of myself as a physician trying to solve problems.”

Special Offer: Free Trial Issue of Forbes