Part IThe Story
The first thing Patrick Soon-Shiong did with the Los Angeles Times, once it was his, was not to walk the newsroom, read the editorial page, or summon the editor to explain the budget. He drove to the printing press. "I wanted to hear the clickety-clack," he said. He had just paid roughly five hundred million dollars for a newspaper most people thought was worth less, and assumed ninety million more in pension obligations. Coming from a man like that, the phrase is so unguarded, nearly a child's phrase, that it is tempting to read it as a put-on. It was not. Or not only.
Half a century earlier and half a world away, on the eastern shore of South Africa, a fourteen-year-old boy had stood at another press, the one that printed the Evening Post in Port Elizabeth. "I still recall the sounds and smells of the printing presses as the first papers rolled off the conveyor belt," he wrote in a full-page letter to readers on June 17, 2018, the day before the sale closed. "I would grab as many as 800 copies from an ink-stained pressman, handing them off to my cadre of 'runners' who would then deliver them to local businesses and residences." He was earning money for university. In the headlines he was also learning, as he put it, "what it meant to grow up 'non-white' under apartheid." The Evening Post closed in 2000. By then the boy who had run its papers through the streets of the North End was a billionaire in Los Angeles, and he was about to lose a five-year fight with his own brother.
That is the shape of the thing, and it recurs. A machine that makes something, and a boy standing near it who does not own it. Later the man owns it, and is not entirely sure what to do with it. Soon-Shiong grew up, he says, surrounded by "a battery factory, a car-tire factory, a meat factory, and the ocean." He made his first fortune by buying two factories nobody wanted. He made his second by wrapping an old drug in a new protein and selling the result, twice, to people who ran bigger factories. In 2021, explaining his COVID vaccine on a digital whiteboard, he drew the virus's nucleocapsid and labeled it the place "where the virus reproduces itself." Then he gave his prescription. "The only way to kill the factory is to have a T cell."
He has been called the richest man in Los Angeles and the world's richest doctor. He has been sued by former investors, by business partners, by other doctors, by his own brother (twice), and by Cher. He has been praised by the Franklin Institute, honored by the Smithsonian, and chastised by the head of the American Diabetes Association. He has been compared by friends to Bill Clinton and by Forbes to P.T. Barnum. He would like, he says, to be remembered "primarily, as a physician-scientist."
In the Smithsonian's oral history of his life, given in 2016, he described a childhood that "felt normal, but it was a strange normal because we were not black, and we definitely were not white. We could get on the bus, but we couldn't sit in the regular compartments with everyone else."
He has spent his life getting on buses. The question his story keeps asking is what happens once he finds a seat.
By the Numbers
The Soon-Shiong Ledger
2 of 200Chinese students admitted to Wits medical school in 1969 under racial quota
800Copies of the Evening Post he grabbed from the press as a teenage newspaper runner
$4.6BPrice Fresenius paid for APP Pharmaceuticals in July 2008
~$2.9BCelgene's 2010 cash-and-stock acquisition of Abraxis BioScience
$147.6M2015 NantKwest compensation package, among the largest of any U.S. CEO
~$500MPrice for the L.A. Times and San Diego Union-Tribune, plus $90M in pensions
$16.6BEstimated net worth in 2026 (Bloomberg and Forbes)
Not Black, and Definitely Not White
His parents were Hakka, from Meixian in Guangdong province, and they left China after the Japanese invasion. His ancestral surname is Huang; his Chinese name, 黄馨祥, is rendered in Mandarin as Huáng Xīnxiáng. The name on his medical degree, his patents, and the masthead belongs to a transliteration of a transliteration, a family identity that had passed through two empires before it reached a third.
In Port Elizabeth his father kept shop. The New Yorker says two small grocery stores; the local Herald says a general dealership in the North End. The accounts also disagree on the size of the family: ten children by one count, eight by another, with Patrick the fifth. They agree about the father's other vocation. In China he had been an herbalist, and in South Africa he became the doctor of the Hakka Association. "Every two months a metal biscuit can would arrive from China filled with foul smelling weeds," Soon-Shiong recalled. "If anybody had a fever, a cough, or a boil, there would be a knock on the door, they'd walk into the house, and he would concoct something. I'd sit there and watch that whole activity." He credits that kitchen pharmacy with a conviction he has carried for fifty years and has lately tried to sell to the FDA: "that you have in your human body the ability to protect yourself from disease. You were given the innate capability to heal yourself."
At home he spoke Hakka and English, with some Afrikaans and Xhosa. He played with Black kids, "what were called 'Coloured' kids," and Indian kids. "There weren't a lot of white kids." He went to a school for Chinese students run by the Anglican Church, where he picked up the cultivated Anglo accent he still speaks in. His science teacher was a priest who had fought in the First World War and been gassed. "He suffered from mustard gas. He could barely talk." So the boy learned science from a man who had lost his voice to chemistry. It is the kind of detail a novelist would cut for being too neat.
He finished secondary school in 1969, at sixteen by Forbes's account, and applied to medical school at the University of the Witwatersrand in Johannesburg. That year Wits registered two hundred students under a strict racial quota: one hundred and ninety-six white, two Indian, two Chinese. To get in he had to be one of the two best Chinese test-takers in the country. He graduated fourth in his class at twenty-three, having delivered, he says, about a hundred babies. While American medical students were memorizing organic chemistry, he was practicing in a Black township. As a nonwhite citizen he was required to carry an identity card at all times.
His wife's family had its own doors to push. Michele Chan was born in East London, an even smaller town up the coast. Her grandfather, as family legend has it, was a sickly infant put out in the pigsty to die because there were already too many children. A servant brought him back inside, saying, "He's not as weak as you think; he's living." The boy left home at fourteen, drifted through India and Southeast Asia, and reached Cape Town as what Michele calls "literally a straggler," arriving in no wave of immigration at all. He opened three businesses. A generation later, when Catholic schools refused Chinese students, Michele's father drove to Port Elizabeth and confronted the archbishop. The archbishop said the exclusion was the law. "Whose law takes precedent?" her father answered. "God's law? To whom do you answer?" Her brother became the first Chinese student to receive a Catholic education there. When Michele enrolled in drama at Wits, the university had to apply to the interior minister for permission for her to study.
So two families, three generations, and the same move each time: find the door, find the man who controls it, and argue. It is not hard to see where Soon-Shiong learned that a rule is something you negotiate with the person enforcing it.
After graduating he was assigned to a "non-European" hospital. He asked to intern at a white hospital in Johannesburg instead, for the better training. Permission was granted on the condition that he work for half pay. He remembers being the only Asian doctor there. An Afrikaans patient with a fever refused to let him touch him and relented only when a supervisor threatened to discharge him. Soon-Shiong diagnosed a sinus infection and had it drained. "So then he's running around the hospital saying, 'You've got to let the Chinaman look at you!'" Soon-Shiong tells the story as comedy, and it is comedy. It is also the template for nearly everything that followed. He is excluded, he is admitted on worse terms, and he wins over the room by being right, loudly enough, in front of the right witness.
There was a darker rotation. In 1976 he was working at a hospital in Soweto when police opened fire on schoolchildren leading the uprising; at least one hundred and seventy-six people were killed. "I was looking after these kids, and now I'm visiting them in the I.C.U.," he said. "That really scared me." One magazine profile says his work in a tuberculosis clinic landed him in police custody. He trained under Bert Myburgh, who had performed South Africa's first kidney transplant and was the rival of Christiaan Barnard, the surgeon who performed the world's first human heart transplant in 1967. Apartheid had isolated South African medicine, and the isolation gave it a swashbuckling style. "South Africa formed me, in a funny way," Soon-Shiong said. "I'm fighting under apartheid, but I'm also being trained by these giants."
He met Michele in his second year of medical school, at a basketball game. In 1977 they left for Canada. The plan, he told the Herald, was to learn the new technologies and come home to build an institute. They never came home. When a reporter later asked whether he had faced discrimination in America, he gave an answer that was half confession and half deflection. "The good news in South Africa was that it wasn't hidden."
You'll Never Survive Academia
At the University of British Columbia he earned a master's in surgery in 1979, with a thesis titled "Fundic Inhibition of Acid Secretion and Gastrin Release." He was the first resident there to win research awards simultaneously from the American College of Surgeons, the Royal College of Physicians and Surgeons of Canada, and the American Association of Academic Surgery. In 1983 UCLA recruited him. He was thirty-one, an assistant professor in gastrointestinal surgery, board-certified a year later. He performed UCLA's first whole-pancreas transplant and became director of its pancreas program.
Then he wanted to shut the program down.
Whole-pancreas transplants carried a high risk of rejection, and he did not think the outcomes justified the danger. He proposed replacing only the insulin-producing islet cells inside the organ. He set up a lab at the Veterans Affairs hospital in West Los Angeles with a staff of three and began sourcing islets from pigs and human cadavers. "The lab was primitive," Kate Hentz said.
Hentz was Lee Iacocca's daughter. In the mid-1980s she toured research labs with her father, who had run both Ford and Chrysler and whose first wife had died of Type 1 diabetes a few years earlier. Iacocca was the archetypal American manager of his era, the man who had saved Chrysler with a government loan guarantee and his own face in the commercials. Now he was looking for a cure, and he had the time and money to shop for one. At the VA he found a surgeon in his thirties with rimless glasses, a long shag haircut, and a soft Anglo-South African accent, who could talk for hours about the human body and then turn to history or literature. The Soon-Shiongs lived in a modest bungalow in Brentwood. "Their house was just adorable," Hentz said. Among its touches were his-and-hers doorframes, cut into the silhouettes of the couple's profiles.
Iacocca agreed to fund the research and urged Soon-Shiong to commercialize it. Soon-Shiong resisted. "He was really excited about what he was doing, but kind of quiet about it," Hentz said. "He was modest. He was humble." Iacocca prevailed, and Soon-Shiong still tells the scene the way you tell a conversion story:
— Patrick Soon-Shiong, recalling Lee Iacocca's advice, to The New Yorker (2021)Lee Iacocca, O.K.? I bring him into my little lab in the V.A. He sits me down and says, "Patrick, you'll never survive academia." I never had any intention of going into business. I wanted to be the chairman of the department of surgery.
Chairman of the department of surgery: the inside seat, the one the quota system had been built to keep him from. He never got it. The rest of the story might be read as an elaborate, very expensive substitute.
Iacocca's money did not end happily either. By the mid-1990s, Hentz said, her father had grown uneasy. Soon-Shiong's relationship with UCLA seemed strained, and the Iacoccas received little information about how their funds were being used. They pulled out. "It was a little odd, our exit, to say the least," Hentz said. "Things weren't totally clear for us." She still calls him brilliant. She also offered the sentence that might stand as an epigraph for everything that follows: "He gets very enthusiastic, and sometimes he might exaggerate. He can embellish a little."
In the early 1990s at least a dozen biotech startups were chasing islet-cell therapies for diabetes, and most of them had the same problem: patients got a few days of remission before their bodies rejected the cells. Soon-Shiong believed he had solved it by encapsulating the cells in alginate, a gel derived from seaweed that is also used to thicken ice cream. In 1992 he implanted pig cells encapsulated this way in a dog. It seemed to work.
In 1993 he found his first human patient. Steven Craig was thirty-eight. Diabetes had wrecked him. He had already had a kidney transplant, he walked with a cane, his eyesight was failing, and he had been out of work for seven years. Soon-Shiong made a two-inch incision in Craig's abdomen and poured in hundreds of thousands of cadaver islet cells wrapped in gel. "It was a simple procedure," he said. "I wouldn't even call it surgery."
A week later, before any long-term result could be known, Craig appeared at a press conference. The CBS Evening News showed him eating a meal "without insulin, for the first time in thirty years." When he had his second procedure that year, the Los Angeles Times, the paper Soon-Shiong would one day own, covered it. "They've done miracles for me," Craig told the paper.
Scott King, who ran a competing islet-cell company, remembered the phone calls. "I had people calling me up saying, 'Diabetes has been cured!' But then he'd take Craig to a conference, and people would ask to see clinical data, and Patrick would skillfully parry them." Craig liked being the star patient. He went to Australia. The family stayed at Beverly Garland's hotel. Soon-Shiong published in The Lancet in 1994, and a later Wits profile would put the result in a single modest clause: the patient stayed off insulin "for an entire month." Then Craig's remaining kidney began to fail, and the media stopped calling. "He grew depressed after that," his ex-wife, Melodie, said. "He liked the attention."
The family told the New Yorker they did not blame Soon-Shiong for what came next. "We understand that sometimes medical practices go wrong," said Craig's stepson, Matthew, now a long-haul trucker. "That's how they find out shit." Michael Zinner, then UCLA's chief of surgery, was less forgiving of the method, if not the man. "He told the world he was curing diabetes," Zinner told Forbes in 2003. "But, in the scientific realm, you need to have your results reproduced by others to have them validated."
Nineteen Ninety-Eight
Three things happened to Patrick Soon-Shiong in 1998. They are usually told separately, and they should be told together.
In 1991 he had left UCLA's full-time faculty to co-found VivoRx, the islet company, with his brother Terrence, a property developer in London. "He was the businessman of the family," Patrick said. In 1994 VivoRx raised five million dollars from the generic-drug maker Mylan Laboratories. That same year Patrick started a second, independent company to develop a chemotherapy drug, and some employees worked for both. VivoRx's investors did not like the arrangement.
On June 29, 1998, VivoRx held a special board meeting in Santa Monica. Patrick says he was asked to sign a document and signed it. By the end of the day he was off the board and the company was suing him. He persuaded Terrence to drop that suit. In 1999 VivoRx, now under Terrence's control, sued him again, this time for fraud, alleging "betrayal, arrogance, greed, and personal aggrandizement that resulted in corporate misconduct of enormous proportions." Patrick denied any misconduct. The brothers settled: Patrick's companies transferred assets worth twenty-four million dollars to Terrence's company, Terrence got the islet-cell patents, and Patrick agreed to stay out of diabetes research for five years.
— Patrick Soon-Shiong, to The New Yorker (2021)That I would not work on diabetes for five years—now that, to me, was evil.
"I said, 'You've hurt mankind,'" he recalled, "because I was this close." Terrence has never discussed the rift publicly. The brothers remain estranged.
This close. The phrase deserves weight, because it returns. Nineteen ninety-eight was also the year Steven Craig, his kidney failing and the attention long gone, checked into a hotel in Orange County and shot himself in the head. Somewhere between the board meeting and that hotel room, the islet-cell era of Patrick Soon-Shiong ended. The man who had been "this close" to curing diabetes had been legally barred from trying, and his first patient was dead.
The third thing that happened in 1998 is the one that made him rich. He bought two pharmaceutical factories, in Illinois and New York, the American generic-injectables operations of Japan's Fujisawa. They had lost money for nine straight years. He had no manufacturing background. Terrence's lawsuit alleged the purchase was made possible by diverting research funds; Patrick denies it and says he financed it independently. Fortune later described him as having "cobbled together loans to buy a struggling generic drugmaker." The plants became American Pharmaceutical Partners. Soon-Shiong returned them to profitability and doubled revenues. In December 2001, APP went public.
It is worth staying with this. The surgeon who could not keep his diabetes company, who had just been sued by his brother and had lost his star patient, took over two money-losing factories that made the unglamorous injectable generics hospitals depend on and that nobody profiles. He made them work. Later he would be accused of many things, but rarely of failing to run a plant. Under APP, his biographies note, the company was responsible for the safe supply of the blood thinner heparin during the 2008 heparin crisis. The factory was the stable thing. Everything else could be argued about.
There is a shelf of hastily assembled paperback biographies of Soon-Shiong now, most of them published in 2025. Some carry titles like Patrick Soon-Shiong: Codebreaker of Modern Medicine: From Billion-Dollar Cures to the Battle for Human Health and THE BILLIONAIRE DOCTOR: Patrick Soon-Shiong's Journey from Apartheid to Global Health Pioneer. As a rule they hurry past 1998. They want the codebreaker and the pioneer. The more instructive figure is the man who, in the year his first dream died, bought a factory.
A Folder Handed Across the Table
The chemotherapy drug was paclitaxel, sold for years by Bristol-Myers Squibb as Taxol and derived from the bark of the Pacific yew. It worked reasonably well. The difficulty was the solvent. Paclitaxel had to be dissolved in a castor-oil product that could provoke allergic reactions, occasionally fatal ones, so patients were premedicated with steroids and infused for hours. Soon-Shiong's idea was to discard the solvent and bind the drug to albumin, a protein the liver makes, packing paclitaxel into nanoparticles about a hundredth the size of a red blood cell. In later years he would describe the concept as "using the tumor's biology against itself." His biographies date the invention variously to 1991, 1995, and 1997. The drug was named Abraxane.
He was a surgeon selling to oncologists. "Patrick is a surgeon by training," said William Gradishar, a breast-cancer specialist at Northwestern. "He was not and is not a medical oncologist. So no one would have known who the hell he was in our world." In the early 2000s, at the annual meeting of the American Society of Clinical Oncology, Soon-Shiong sat down with Gradishar and asked him to run a clinical trial. Then he did something strange.
"He handed me a manila folder," Gradishar said. "And he goes, 'I want you to read this. You're going to hear some stuff about me, so here it is.'" Inside was a magazine article about the lawsuits his brother had filed against him. "You know, that was a little bit strange," Gradishar said. "But at least he was up front about it."
The folder is one of the most revealing objects in the story. It reflects a man who has learned since childhood that the case against him will be made whether or not he is in the room, and who has concluded that the best move is to make the room his. Here is the worst thing you will hear about me. Read it now, on my terms, across my table. Gradishar was wary but impressed by the data and by the oncologists Soon-Shiong had assembled, and he agreed to serve as principal investigator.
He had walked, without knowing it, onto a Wall Street battlefield. The U.S. rights to Abraxane belonged to APP, the public company. The world rights and the patent belonged to a private company, American BioScience, of which Soon-Shiong owned eighty percent and which was itself APP's largest shareholder. The structure enraged some investors and attracted short sellers. Analysts questioned the trial design: Soon-Shiong had fired the American firm running the trial and finished it in Russia. At one point nearly all of APP's available shares had been sold short. Forbes ran a piece in October 2003 headlined "Vindication," which asked why so many people wanted to see him fail and quoted him saying, "This is the culmination of my career's obsession." Meanwhile investors bid APP's market value up 250 percent in a year, to $2.4 billion. Through his stake in American BioScience, Soon-Shiong owned about fifty-two percent of APP, and his net worth rose from $390 million at the IPO to $1.1 billion.
The trial enrolled four hundred and fifty-four women with breast cancer. Tumors shrank in thirty-three percent of those who got Abraxane and nineteen percent of those who got standard paclitaxel, which works out to about thirty-four more women responding. Survival was not much better. By some measures the paclitaxel arm did worse than paclitaxel had in other studies, which may have flattered the comparison. But the tumors had shrunk.
In early 2005, against the short sellers' expectations, the FDA approved Abraxane for metastatic breast cancer. APP's stock rose forty-seven percent. Soon-Shiong commissioned a commemorative paperweight displaying the stock chart.
He had been formed in a country where the score was kept in public, and he kept score in glass.
Old Wine, New Bottle, Same Owner
"It is the nature of the American health-care system," Stephen Witt wrote in the New Yorker, "that marginal improvements can result in vast fortunes." An independent review in Annals of Oncology in 2006 concluded that Abraxane and drugs like it did "not really" offer a significant therapeutic advantage over established medicines, and called them "old wine in a new bottle." Harold Burstein of Dana-Farber was blunter in Fortune in 2013: "It has a slightly different toxicity profile. For some patients it's a nice trick to know about. But in terms of its benefit in breast cancer, there is none."
That was the clinical view. The commercial view was simpler. Abraxane now had a billing code. A 2006 New York Times article reported that it sold for $4,200 a dose, twenty-five times the price of generic paclitaxel; Soon-Shiong says he had thought the cost was much lower. Doctors who administer such drugs may receive a percentage of the price. "The incentives were exactly backwards," said Peter Bach of Memorial Sloan Kettering, who has spent his career fighting drug prices. Bach is also Soon-Shiong's friend. "I've been on his plane. I've gone skiing with him. Knowing Patrick has enriched my life considerably."
That pairing of severe criticism and genuine affection, often from the same person, recurs so consistently in the reporting on Soon-Shiong that it begins to look less like a contradiction than a description. Jennifer Malin, the oncologist who designed Anthem's 2014 program to pay doctors for prescribing cost-effective cancer treatments, excluded Abraxane for breast and lung cancer and kept it only for pancreatic. She remembered a meeting: "You go into the conference room, and he just talks for like three hours straight and fills up this giant whiteboard with all his theories of the way things work, whether or not they're based in reality." A non-oncologist, she said, might conclude he was just brilliant. "You really have to be an oncologist to be able to say, 'You know, this stuff is kind of wacky.'" She also remembered a dinner at which his wife and two children were eating on the other side of the restaurant, and he kept going over to spend fifteen minutes at their table before returning to hers. "If I was married to him, I'd be livid! Anyway, that's pretty funny. He's like an excited kid. It's hard to get upset."
The drug kept improving on paper. The FDA added lung cancer in 2012 and pancreatic cancer in 2013, in combination regimens, and his own biographies describe Abraxane as the first drug in more than two decades to show a significant survival improvement in metastatic pancreatic cancer. In 2019 about eighteen thousand Medicare beneficiaries received it, at average spending above seventeen thousand dollars each; paclitaxel cost Medicare about a hundred and thirty. Annual sales passed a billion dollars. In 2017 the Smithsonian put Abraxane on display in the National Museum of American History, the same year it invited him and Michele into its permanent exhibit on immigrant journeys.
The corporate history of these years reads less like strategy than like a man folding and unfolding the same piece of paper to see which crease holds. On November 28, 2005, months after approval, APP announced it would merge with American BioScience, its own parent and largest shareholder, which then owned about 64.4 percent of APP. Based on APP's stock price, the combined company would be worth about $7.5 billion. The deal closed on April 18, 2006. American BioScience shareholders received 86,096,523 new shares on top of the 47,984,160 they already held, and the stock began trading as Abraxis BioScience, ticker ABBI. Soon-Shiong now owned more than eighty percent of the company he had taken public a few years earlier. "We believe this merger will allow us to maximize the true potential of both organizations as one fully-integrated, global entity," he said.
On July 2, 2007, less than fifteen months later, he announced that the company would split in two again. The shares had lost nearly a quarter of their value since the merger. "By separating these unique business units," he said, "we believe we will be able to unlock the intrinsic value of both companies." The SEC filing described the logic: the generics business, renamed APP Pharmaceuticals, would raise debt on better terms and contribute about $725 million of the proceeds to the proprietary business, which kept the name Abraxis BioScience and the ticker ABII. The separation was completed in November 2007. In July 2008, Fresenius bought APP for $4.6 billion. In 2010, Celgene bought Abraxis for roughly $2.9 billion in cash and stock (higher figures, which count contingent payments, also circulate). Soon-Shiong became Celgene's largest individual shareholder, and over the next four years Celgene's stock tripled.
Forbes reporter Matthew Herper found him in October 2009, a year before the Celgene sale, in a dim basement conference room at the Peabody hotel in Orlando. Over a breakfast buffet, Soon-Shiong was trying to persuade a dozen oncologists to enroll pancreatic-cancer patients in an Abraxane trial. The drug was selling about $300 million a year, a tenth of what blockbusters made. Abraxis had lost $278 million in 2008, $159 million of it to unwind a marketing partnership with AstraZeneca, and its stock was down sixty percent for the year. The only analyst covering the company, Michael King of Merriman Curhan Ford, summarized the view from the Street: "I think he's viewed as 'What's best for Patrick is what carries the day.'"
Soon-Shiong changed course mid-pitch and told the doctors about an idea that had been "whirling in my head for two years," a theory of metastasis in which Abraxane turned out to be exactly the right weapon. The doctors were energized, if not convinced. "The hypothesis is a great hypothesis," one said. "We need the clinical data to back it up." Soon-Shiong was unbothered. "If this is real," he said, "it is profound."
Herper identified the mechanism of the fortune in a single line, and it was not science: Don't give up equity. Soon-Shiong had kept ownership of the drug he invented and given away little of the generics business. He was, Forbes noted, the only drug researcher who had become a billionaire. Years later, asked about his reverse mergers, he would put the method in his own terms.
"The reverse merger has nothing to do with money or stock," he said. "It's to do with putting the right ingredients into the right mixing bowl. So you can cure patients."
He had started with nothing fourteen years earlier, working outside his specialty, and he was now worth more than seven billion dollars.
Everything Was a Circle
"To me, as a kid, everything was a circle—there's no beginning and no end," he told the New Yorker. "And what I mean by that, as a systems engineer, is really looking at integrating—connecting dots."
With cash from the sales, he set out to connect everything. In 2009 he told CNBC he would spend a billion dollars building a "smart grid" for medical information, which he called "a Bell Labs of healthcare," "a public utility," and "a medical information superhighway," a cure for the proprietary record systems he dismissed as "medical bridges to nowhere." Asked whether insurers supported the idea, he said, "Not really." He laughed about the doubters. "That's the most exciting time in life, when people tell me I can't do that. I don't know what it is about me, when you say, 'You can't do that,' because then I say, 'Well, that's great, because then that's exactly what I should do.'"
He bought the National LambdaRail fiber network in 2011. By 2012, Reuters reported, he had spent more than $400 million of his own money on fiber linking cancer clinics nationwide. Promotional material described a 12,000-mile network and algorithms that could analyze a human genome in forty-seven seconds. "This is something the federal government should have done, but we waited and waited for them," he said. In a warren of offices on the Westside, floor-to-ceiling flowcharts mapped the plan, known internally as "The Rocket Ship." NantHealth was founded in 2007. NantWorks, the umbrella, followed in September 2011, with a mission statement that read like a list of nouns: "to converge ultra-low power semiconductor technology, supercomputing, high performance, secure advanced networks and augmented intelligence to transform how we work, play, and live."
The NantWorks logo is a feather coming out of a circle. When Witt asked what "Nant" meant, Soon-Shiong pointed to an Apache basket on his shelf. "The word 'Nantan' stands for 'he who speaks for the people,'" he said. "I'm an honorary Navajo, and I'm on the Apache council. Because my job, frankly, is to help the marginalized and underserved." Asked the same question later by Cancerworld, he let the meaning drift: "the word could have whatever meaning you want." It could stand for "neural artificial net." It could be "the neural ant, because ants follow each other through signals." It could be nanotechnology. A name able to mean anything fits a holding company that eventually contained NantHealth, NantOmics, NantKwest, NantCell, NantPharma, NantMobile, NantEnergy, NantGames, NantStudios, NantCloud, and NantRenewables, along with zinc-air batteries, graphene, bioplastics, and a "next generation urban scooter."
Some of it worked. In 2013 he invested in a video-conferencing startup called Zoom, then valued at fifty million dollars. Eight years later it was worth seventy billion.
Much of it was louder than it was successful. In July 2015 NantKwest, a natural-killer-cell company, went public at a market value of $2.6 billion, at the time the highest-valued biotech IPO ever. It priced at $25, opened at $37, and fell more than eighty percent. Soon-Shiong's 2015 compensation from NantKwest was $147.6 million. NantHealth went public in June 2016 at $14 a share, got a Nasdaq delisting warning in January 2019 when its market value was about $57 million, and did a 1-for-15 reverse split in December 2022.
On January 12, 2016, he launched the Cancer MoonShot 2020, a coalition of companies, insurers, and academics that would put a cancer vaccine through trials in twenty thousand patients by 2020. He pitched it to Joe Biden, discussed it with Bill Clinton, and talked about it with the pope. "We have lost the war on cancer," he told a Scripps conference that March. STAT investigated a year later and concluded that the moonshot functioned largely as marketing for GPS Cancer, a costly tumor-sequencing test sold by one of his companies, and that several announced "milestones" amounted to hospitals agreeing to use the test. "The clinical breakthroughs touted by Patrick Soon-Shiong are less than modest—they are the most minuscule and vague findings," said Vinay Prasad, an oncologist then at Oregon Health and Science University.
Then came Utah. In 2014 Soon-Shiong and his foundations gave the University of Utah $12 million for genetic research, and the university paid $10 million of it to NantHealth. In November 2016 NantHealth told investors it had received 524 orders for GPS Cancer in a quarter, and Soon-Shiong said on the earnings call that a third came from the Utah deal. When STAT reported this on March 6, 2017, NantHealth shares fell thirty-one percent in two days. Soon-Shiong called the story "maliciously false." "I'm befuddled by any other motivation that could be perceived," he said. A state audit found that the university had broken procurement law by skipping competitive bidding. Utah's House speaker, Greg Hughes, said the arrangement looked like an attempt "to Cinderella-slipper something for one person, or for one entity." Politico found that the Chan Soon-Shiong NantHealth Foundation had spent more than seventy percent of its money on businesses and nonprofits he controlled.
In May 2015 NantPharma paid $90 million for IgDraSol, the Sorrento Therapeutics subsidiary that owned Cynviloq, another protein-bound paclitaxel. The contract promised another $1.2 billion in milestone payments, including one for FDA approval. In South Korea, where Cynviloq competed with Abraxane, Abraxane sold for $270 a dose against Cynviloq's $180; in the U.S., court documents put Abraxane at $1,378 a dose, with no direct competition. Sorrento alleges that after the purchase Soon-Shiong abandoned the approval process and let key patents lapse, a tactic known in the industry as "catch and kill." Soon-Shiong said in a statement that the suit was "a cynical attempt to deflect from Sorrento's own breach of contract." He argued that Sorrento had manipulated data and that the drug had manufacturing problems. "Just because someone's charismatic, and just because they're a billionaire, that doesn't mean what they're saying is true," said Steve Feldman, Sorrento's lawyer. Cynviloq is still not available in the United States. Cher sued him in 2017, alleging that she had been misled into selling shares in a promising drug company; her case was dismissed in 2018.
The circle closed in a way no one could have planned. In 2017 his companies took control of Verity Health, which ran six California hospitals. Verity filed for bankruptcy in 2018, after spending more than twenty million dollars on an IT upgrade from a vendor in which he held a stake. He then took control of St. Vincent Medical Center, near downtown Los Angeles, the hospital where he had operated on Steven Craig. It failed as well, and the building became a temporary COVID ward. "The nuns hadn't funded their pensions," Soon-Shiong said.
Witt, near the end of his reporting, named a principle after him. "Ask him a question about medicine and you will receive an answer about business; ask him a question about business and you will receive an answer about medicine; but rarely will you receive both answers at the same time." He called it the Soon-Shiong uncertainty principle. It sounds like a joke, and it is also a fairly exact description of how a man with no single home domain keeps from being pinned down in any of them.
Courtside
In 2010 he bought Magic Johnson's 4.5 percent of the Los Angeles Lakers, which came with four courtside seats, the most visible perch in a city that runs on being seen. Even so, he was described in 2012 as one of the least known of the city's billionaires. "He's not one of those owners who wants to be seen everywhere," Kobe Bryant told Reuters. "He's just one of the fans." Bryant hugged him before every game, for luck "and maybe some of the success."
In 2012 he joined a failed bid for the Dodgers. He also made a run at Philip Anschutz's AEG, and fell short. Meanwhile he kept buying the lots around the Brentwood bungalow, twelve parcels in the end, and built a compound his friend Burt Bacharach preferred to call "a campus." Under the living room is an NBA-regulation basketball court lit by natural light, through a design on which Soon-Shiong holds three patents. "This court is the best court I've ever seen in my life," said Metta Sandiford-Artest, who played nineteen NBA seasons as Ron Artest and then as Metta World Peace. "It's insane. It's deep under the floor... with locker rooms and televisions. And bowling alleys." He also admired the owner's game: "He can pull up left, pull up right, he has a one-dribble fade."
"To this day, basketball is the thing that keeps me sane," Soon-Shiong said. "It's the only activity where I really truly relax." Asked by Cancerworld to describe himself in a sentence, name a role model, and state his dream, he gave the same answer three times: "I wish I could have been Kobe Bryant."
Asked for his close friends, he named three: Bacharach, Sandiford-Artest, and Jerry Zucker, who produced Airplane! "Patrick is brilliant," Zucker said. "I find him fascinating, on the rare occasion that I understand what he is talking about." Bacharach, at ninety-three, told a story about his son Oliver, hospitalized with an antibiotic-resistant staph infection. Soon-Shiong, who was not on staff, came by, asked to see the chart, spoke gently with the infectious-disease doctor, and suggested a different antibiotic. "The next day, Oliver was better."
That is the doctor at his best: an outsider who gets into the room by being gentle and then turns out to be right.
The same move can also be seen in the Bryant episode. On April 12, 2013, Bryant ruptured his Achilles tendon during a Lakers game. Soon-Shiong, who had ruptured his own Achilles playing basketball and believed he had developed a better approach, went to the locker room and urged immediate surgery rather than the usual wait for swelling to subside. About forty-five minutes after the injury, the orthopedist Neal ElAttrache's home phone rang. "It was Patrick, in the locker room with Kobe." ElAttrache is known for being sober and evidence-driven, and he was scheduled to operate on the pitcher Zack Greinke the next morning. At Soon-Shiong's urging he did both surgeries back to back. Soon-Shiong, who had not operated in years and had no orthopedic training, was in the room.
The two men tell the story differently. Soon-Shiong: "The treatment for Achilles rupture is completely wrong. I asked Kobe if he wanted to dunk again, and he said yes. So we went completely against the doctor's orders, and Kobe has his treatment, and he dunked again." ElAttrache: "That this was some sort of novel thing that no one else had considered, you know, I just, I don't know... I think that's a little bit of an overstatement." He said Soon-Shiong's input had not changed his technique. He also said this: "Patrick functions on the edge. You need people like that. I ask myself, 'Is there some kernel of genius in there that can help the people I need to take care of?' So I listen to him. I definitely listen."
Bryant played again. He never won another championship. "He wasn't a hundred per cent after that," Sandiford-Artest said. "No way."
Proximity was the real asset. The locker room led to the operating room. Courtside led to Iacocca, Clinton, Biden, and Gates. In late 2016 he and Michele gave a hundred thousand dollars to Hillary Clinton's campaign; on November 19, twelve days after the election, he was photographed arriving at Trump National Golf Club in Bedminster. Politico reported that he was seeking a cabinet post, and the Boston Globe that he had pitched himself as a national health czar. He did not get a post. In May 2017 Speaker Paul Ryan appointed him to a federal health-IT advisory committee. Scott King, the old islet-cell rival, described the effect of his attention: "He's like Bill Clinton. When Bill Clinton is talking to you, you get the feeling that you're the only person in the world. He gives you the impression that he's instantly there for you, all out, even though you know that can't possibly be true."
Soon-Shiong describes the pleasure from the other side, as the one doing the knowing. "Knowing, even if just for an instant, that you are the only person in the world who understands this idea or this result," he told Cancerworld, "that is what drives me."
The Trophy
The Los Angeles Times was founded as an instrument of exclusion. Around 1900, Harrison Gray Otis, a former Army general, used it to promote Los Angeles as a paradise for the "Anglo Saxon" ideal. His son-in-law Harry Chandler, the man behind the Hollywood sign, ran it much like a real-estate circular. Labor activists bombed its offices. In the 1960s it moved left with the rest of California, and in the 1980s, at its peak, it won Pulitzers, carried a thick classified section, and filled a city block downtown. By 2018 it had been sold to Chicago owners, its Art Deco headquarters had been sold, its newsroom had been cut roughly in half, and plans were circulating to cut up to twenty percent more and close the Washington bureau.
Soon-Shiong said that last prospect is what made him move "as desperately fast as possible." He said something else, too, which is less often quoted. In 2007 the Times reported that a forty-three-year-old woman had lain dying on the emergency-room floor at Martin Luther King Jr.-Harbor Hospital while her boyfriend called 911 from a pay phone outside. "That story," Soon-Shiong said, "it haunted me." He drove to the hospital, found the doctor who had been in charge that night, and told him, "This is completely unacceptable." His foundation later provided a hundred-million-dollar guarantee to help rebuild the hospital. "I wouldn't have known had it not been in the paper," he said.
He came in through a side door. On May 23, 2016, while Tribune Publishing was fighting off a hostile bid from Gannett, his Nant Capital bought 4.7 million shares at $15 each, or $70.5 million, the same per-share price Tribune had called "grossly inadequate" when Gannett offered it. He became the No. 2 shareholder and an ally of chairman Michael Ferro. The alliance soured. On February 7, 2018, he agreed to buy the Times, the San Diego Union-Tribune, and several smaller Southern California titles for about $500 million in cash plus $90 million in pension obligations. That was twice what Jeff Bezos paid for the Washington Post, which had three times as many subscribers. Eli Broad, David Geffen, and Austin Beutner had all circled the paper for years. The New York Times said friends regarded it as the ultimate prize of social cachet, and noted that the buyer lived next door to Beutner.
That evening he sent a letter to the staff. "My own family immigrated from southern China to South Africa generations ago. We chose to settle in Los Angeles because this is the place that most felt like home... As someone who grew up in apartheid South Africa, I understand the role that journalism needs to play in a free society." The full-page letter to readers in June went further, and in his own idiom: "I believe that fake news is the cancer of our times and social media the vehicles for metastasis." He called the purchase "the culmination of an immigrant's American Dream." He and Michele were the first nonwhite owners in the paper's history. Otis would have hated it, which was arguably part of the appeal.
He hired Norman Pearlstine, who had run Time Inc.'s editorial operations, as executive editor. Soon-Shiong later said he had "begged" him. He stopped the layoffs and added more than a hundred and fifty people to the newsroom. He announced that the paper would leave the downtown Art Deco building, its home since 1935, and move to El Segundo, a sixteen-thousand-person rectangle of tract housing hemmed in by LAX, an oil refinery, and a sewage plant, which also served as the sandbox for NantWorks. Michele designed the new space, including a large test kitchen for expanded food coverage.
— Norman Pearlstine, former executive editor of the Los Angeles Times, to The New Yorker (2021)He made the acquisition with very little due diligence, because he thought that it had to be easier than curing cancer. I'm not sure whether he still believes that.
Then the pandemic came. "We would speak several times a week, occasionally several times a day, until March of 2020," Pearlstine said. "We had far less communication from that time on." Soon-Shiong took COVID as seriously as anyone Witt had met. He rarely left the compound, his PR representative had not been in a room with him since March 2020, and he would not meet a reporter in person. Asked which vaccine he had received, he said, "I can't tell you that," and then implied he had used his own experimental one. Kevin Merida, a former Washington Post managing editor who had run ESPN's Undefeated, became executive editor in June 2021 and did not meet the owner in person until September. Asked how often they talked, he said, "We don't have a cadence."
In the meantime, after George Floyd's murder in 2020, the Guild's Black Caucus wrote him a public letter, and he answered it. He hired more Black journalists, commissioned an apology for the paper's racist past, and wrote an op-ed: "As the Los Angeles Times' first nonwhite owners in its nearly 139-year history, my wife, Michele, and I are determined to increase diversity." Erin B. Logan, the caucus's twenty-six-year-old chair, was moved. "Newspapers tend to be owned by a certain kind of person. But Patrick is different. He's personally experienced discrimination like this."
The paper had about four hundred thousand subscribers in 2021, against eight million at the New York Times. Asked how much he had spent in total, he told the Press Gazette, "We may be close to $1bn by now. But I don't have a fixed number because I'm, to be honest with you, not really clear." Asked whether the paper was now worth more than he paid, he laughed. "Most people would say it wasn't worth $500m to start with, right?" Before Witt's first interview, his office sent a one-page sheet listing twenty-five business lines. The Los Angeles Times was twenty-third, alongside the bioplastics company, the cloud venture, the water purifier, and the scooter. When Alden Global Capital, the hedge fund a columnist called the "destroyer of newspapers," bought Tribune Publishing in 2021, Soon-Shiong did not oppose it. "Local benefactors should manage local papers," he said. "I couldn't manage the Baltimore Sun."
A company spokesperson told the New Yorker that the test kitchen was "ninety per cent done."
Fair and Balanced
He had owned the press for years without much touching the copy. Pearlstine said he never interfered with coverage. The opinion page was a different matter. In 2020 Soon-Shiong blocked the editorial board from endorsing Elizabeth Warren in the Democratic primary, though the paper endorsed Biden in the general election. His daughter, Nika, a doctoral candidate at Oxford who ran a guaranteed-income program in Compton, took a growing interest in the paper. When the Wall Street Journal reported in February 2021 that he was considering a sale, she tweeted, "WSJ is 100% wrong." Staffers came to see her contacts with them, public and private, as meddling. "A light switch has gone off for me," she told Witt, "of understanding the influence that public perception and mass narratives have over public-policy decisions."
In October 2024 the editorial board was preparing to endorse Kamala Harris, who had been a senator from California. Soon-Shiong blocked it. It was the first time since 2004 that the paper had not endorsed a presidential candidate. He said he had asked for a "factual analysis" of both candidates' policies instead. TheWrap reported that a planned series, "The Case Against Trump," was also killed. Mariel Garza, the editorials editor, resigned. "I want to make it clear that I am not OK with us being silent," she said. Robert Greene, a Pulitzer winner, and Karin Klein followed her, and so did the legal columnist Harry Litman, who called his resignation "a protest and visceral reaction against the conduct of the paper's owner, Dr. Patrick Soon-Shiong." By the Times's own later estimate, about twenty thousand subscribers cancelled.
He did not retreat. In November the paper dismissed the editorial board, and he promised a "fair and balanced" newspaper, borrowing Fox News's old slogan, and a "rebirth" in which "every American's views should be heard." In December he announced an AI-powered "bias meter." In January 2025 a psychiatrist named Eric Reinhart said his op-ed opposing Robert F. Kennedy Jr.'s confirmation as Health and Human Services secretary had been cut, without his approval, so that it no longer opposed it. His proposed headline, "RFK Jr's Wrecking Ball Won't Fix Public Health," became "Trump's healthcare disruption could pay off — if he pushes real reform." Soon-Shiong posted it on X, calling Kennedy "our best chance." On March 3, 2025, he introduced "Voices," a label for opinionated content, and "Insights," an AI feature that places each Voices piece on a political spectrum and offers "different views on the topic." That same month he told Tucker Carlson that COVID, and the vaccines that had not stopped it, were likely driving an epidemic of aggressive cancers. This came from a man who had built his own COVID vaccine and, by his own suggestion, given it to himself first.
There is a reading of all this as capitulation to a new administration. There is a reading of it as a sincere, if clumsy, effort to restore balance to a paper he called an "echo chamber." There is a third reading that the record supports at least as well as either. A man who has spent his life on the edge of every category, not Black and not white, not an oncologist and not a pure businessman, not quite inside and never fully excluded, now owned an institution that had always told Los Angeles which side it was on. Asked to pick one, he declined. In medicine that refusal had made him rich. In the opinion section it lost twenty thousand subscribers. "I'm extremely proud of work we've done right," he said in December 2024, citing six Pulitzers during his ownership, "and we've done a lot right."
The newsroom shrank. Two rounds of layoffs erased most of the hiring that followed his purchase. In July 2023 he sold the Union-Tribune to MediaNews Group, an Alden affiliate. In October 2025 he began offering up to $500 million in private stock, the same amount he had paid for the paper, ahead of a planned listing on the New York Stock Exchange under the ticker LAT. Investors were offered a seven percent dividend and a twenty-five percent conversion discount, and he would keep the voting control. The 249-page placement document included the sentence: "We will have broad discretion in the use of proceeds from this Offering and may not use them effectively." He described the IPO as a way to "democratize" the paper. The pitch read: "We'll write the next 144 years, together."
The boy who once handed papers to runners in Port Elizabeth was now trying to sell pieces of the press to the readers.
Ninety Per Cent Done
Through all of it, the doctor's original conviction held, the one formed watching his father mix weeds from a biscuit tin. You were given the innate capability to heal yourself. By 1990, he has said, he had identified the natural killer cell as the key immune cell in both organ rejection and cancer killing: "nature's first responder." NantKwest had been built on an NK cell line. ImmunityBio, founded in 2014, merged with the public NantKwest in March 2021 in another reverse merger, which again left Soon-Shiong with more than eighty percent of the surviving company. In 2021 a Wedbush analyst said ImmunityBio might be three to five years from a marketable product. "I don't have any questions from investors on it," he said. "No one. No one cares."
In September 2021 Soon-Shiong appeared alongside President Cyril Ramaphosa to announce NantSA, a vaccine venture for sub-Saharan Africa. In January 2022 they opened a manufacturing campus in Cape Town together, with investment reported at more than four billion rand. The surgeon who had left South Africa in 1977 meaning to come back and build an institute had, forty-five years later, come back and built one.
In February 2022 ImmunityBio reported that an IL-15 superagonist, N-803, combined with BCG had produced complete remission in seventy-one percent of patients with BCG-unresponsive non-muscle-invasive bladder cancer, with a median duration of more than two years. In April 2024 the FDA approved it as Anktiva. It was the first approval for his NK-cell thesis after thirty-four years. Approvals in the European Union, the United Kingdom, and Saudi Arabia followed. In June 2025 the FDA granted expanded access for patients with lymphopenia and solid tumors. ImmunityBio's market value reached about eight billion dollars, his stake roughly 62.5 percent. His net worth reached an estimated $16.6 billion. Sixty-five years after the Wits quota, it would be hard to argue he had not been, finally, this close.
In March 2026 the FDA sent ImmunityBio a warning letter. The agency said claims made on a podcast and in a television ad overstated Anktiva by presenting it as a "cancer vaccine" for multiple solid tumors without adequately disclosing risks. The company's website now opens with a banner: Reaffirming our Commitment to Advertising Compliance. Read our Corrective Communication here. Thirty-three years after the press conference with Steven Craig, the pattern repeated: the result was real, and he described it as larger than it was.
Near the end of his interviews with Witt, Soon-Shiong was asked a question he seemed never to have considered. What qualities had allowed him to succeed in medical school, under openly declared white supremacy? For a moment the man the Iacoccas had met in the VA lab reappeared, modest and unsure. "Well, I have a good memory, right?" he said, after a time. "I think I was given a little bit of a gift, that I can see things in a different way."
In Brentwood the compound now covers twelve parcels. Under the living room is the court where Kobe Bryant used to play, lit by daylight through a design protected by three patents. The original bungalow, the modest one Kate Hentz found so adorable, is the guesthouse now. Its doorframes are still there, cut to the outlines of two Chinese South Africans who were once not allowed to sit in the regular seats on the bus: one shaped to his profile, one to hers.
Part IIThe Playbook
What follows is not a recommendation to become Patrick Soon-Shiong. It would be unwise, and in some respects inadvisable on legal grounds. It is an attempt to separate the transferable mechanics of his career, the moves that built the fortune and the moves that eroded the trust, from the mythology he and his admirers have built around them. Some principles are things to copy. Some are warnings stated as principles. A few are both. The arc runs from a half-pay internship in Johannesburg to a $16.6 billion fortune, and nearly every step in it contains one decision that compounded and one that cost him.
The principles
- 1.Never sell the thing you invented.
- 2.Fund the moonshot with a factory.
- 3.Treat corporate structure as a reversible instrument.
- 4.Hand them the folder first.
- 5.A marginal advantage plus a billing code is a business.
- 6.Narrative is capital, and it accrues interest in both directions.
- 7.Answer in the other language.
- 8.Proximity precedes the deal.
- 9.Every related party is a future headline.
- 10.Your incentives will be read as your intent.
- 11.Attention is the owner's scarcest asset.
- 12.Stay on the thesis long enough to be right.
Principle 1
Never sell the thing you invented
Matthew Herper's 2009 diagnosis in Forbes still holds: Soon-Shiong "has worked an age-old formula for getting rich: Don't give up equity." Through every phase of Abraxane, the short sellers, the Russian trial, the merger and the split, he owned the patent through a private entity he controlled at eighty percent. Biotech founders typically dilute themselves to irrelevance in exchange for venture capital and big-pharma partnerships. Soon-Shiong did the opposite. The Los Angeles Times noted in 2016 that the drug "may never have made it to market without Soon-Shiong's aggressive business style and his insistence that he remain in charge of his idea—a rare accomplishment in the pharmaceutical industry."
The price of that control was isolation. One analyst covered the stock, and he described the company as run on the principle of "what's best for Patrick." Investors complained about the structure in public. But when Celgene paid roughly $2.9 billion, most of the money went to one person.
Tactic: Before raising money, identify the one asset whose ownership determines the outcome, and structure every financing so that someone else's capital funds the periphery while you keep the core.
Principle 2
Fund the moonshot with a factory
The overlooked move of 1998 was buying two money-losing injectable-generics plants that had been in the red for nine straight years. Soon-Shiong had no manufacturing experience. He had a drug that needed years of trials and a reputation that made venture investors hesitant. The factories solved both problems. They produced cash, they gave him a public vehicle (APP's 2001 IPO), and they gave the speculative asset a respectable home. Fortune's summary: he "turned the company around, used the profits to develop Abraxane, preserved his equity, and cashed out a decade later."
In the end the boring business was worth more than the glamorous one: $4.6 billion for APP against roughly $2.9 billion for Abraxis. The factory was not merely a means of financing the vision. It was the larger part of the fortune.
Tactic: If your core bet has a long and uncertain path, look for a distressed, cash-generative operating business in an adjacent supply chain that you can fix, and let it pay for the research without selling equity.
Principle 3
Treat corporate structure as a reversible instrument
Soon-Shiong merged APP into its parent in April 2006, split the combined company apart in November 2007, sold the two halves in 2008 and 2010, and repeated the method in 2021 with NantKwest and ImmunityBio. Each move had a stated rationale: full integration, then unlocking "intrinsic value," then better debt terms and $725 million moved into the proprietary business. Each also served a less-stated purpose: keeping control while changing what the market was pricing.
⟲
The Accordion
Fold, unfold, sell: how one founder restructured the same assets repeatedly while keeping control.
1998
Buys two money-losing generic-injectables factories; they become American Pharmaceutical Partners.
2001
APP goes public; Abraxane world rights stay in private American BioScience, 80% owned by Soon-Shiong.
2005
FDA approves Abraxane; APP stock rises 47%. On Nov. 28, APP agrees to merge with its own parent.
2006
Merger closes April 18 as Abraxis BioScience (ABBI); Soon-Shiong owns more than 80%.
2007
Split announced July 2 after shares fall nearly 25%; completed in November as APP and Abraxis.
2008
Fresenius acquires APP for $4.6 billion.
2010
Celgene acquires Abraxis for roughly $2.9 billion; he becomes Celgene's largest individual shareholder.
2021
Public NantKwest merges with private ImmunityBio; he again holds more than 80%.
2024
ImmunityBio's Anktiva wins FDA approval.
The "mixing bowl" line is partly a dodge. It is also an accurate description of a mental model in which legal entities are containers to rearrange, not identities to defend. The cost is credibility. Each reshuffle that benefits a controlling shareholder makes the next one look suspect.
Tactic: Design your holding structure so that integration and separation are both available later, and tie each restructuring to a measurable capital or operating outcome so that it reads as strategy rather than self-dealing.
Principle 4
Hand them the folder first
Soon-Shiong's first meeting with William Gradishar, the oncologist who would run the trial behind his fortune, began with a manila folder containing the worst press about him. "You're going to hear some stuff about me, so here it is." Gradishar thought it was strange. He also thought it was honest, and he signed on.
This is reputation management as a preemptive strike. Disclosure turns a hidden liability into proof of candor, and it lets you frame the story before a rival does. Soon-Shiong learned under apartheid that the hostility was out in the open ("the good news in South Africa was that it wasn't hidden"), and he seems to have concluded that hostility you name yourself is easier to manage than hostility you leave hidden.
Tactic: Before any high-stakes partnership, give your counterpart the strongest version of the case against you, in writing, along with your answer, so that they hear it from you first.
Principle 5
A marginal advantage plus a billing code is a business
The clinical record on Abraxane's first indication was modest: a response rate of thirty-three percent against nineteen, similar survival, and a review calling it "old wine in a new bottle." The commercial fact was decisive: approval, a reimbursement code, and a price twenty-five times the generic's. Once the drug was in the system, adding lung and pancreatic cancer, plus easier administration and no steroid premedication, built a franchise that eventually did more than a billion dollars a year.
The lesson is uncomfortable, and that is why it is worth stating. In markets with administered pricing, regulatory status can be worth more than the size of the improvement. What Soon-Shiong saw was not that the albumin wrapper was revolutionary. It was that any defensible improvement, once approved and coded, is priced by the system rather than by its marginal value. Peter Bach's verdict, that "the incentives were exactly backwards," describes the opportunity and the indictment equally.
Tactic: In regulated markets, map where an approval, certification, or code creates pricing power, and ask whether a modest but defensible improvement to an existing product can earn it faster than a new invention could.
Principle 6
Narrative is capital, and it accrues interest in both directions
Scott King judged it fairly decades later: "At the time, I found this kind of self-promotion unethical. But I will say, now that I'm older, I realize you need someone like that on your side. He was good at raising money." Steven Craig's press conference one week after surgery, the MoonShot's twenty-thousand-patient promise, "quantum oncotherapeutics," the "largest biotech IPO in history" all brought in capital, partners, and presidents.
The same habit runs through every setback. It produced the American Diabetes Association's rebuke, Zinner's "He told the world he was curing diabetes," STAT's moonshot investigation, and the 2026 FDA warning letter over Anktiva's promotion. Even real results got discounted, because he had described smaller ones in the same superlatives.
§
Promise Versus Record
What was announced, and what followed.
| Venture | The claim | The record | Verdict |
|---|---|---|---|
| Islet cells (1993) | Patient eats "without insulin, for the first time in thirty years" | About one month off insulin; ADA rebuke; patient's kidney failed | Overstated |
| Abraxane (2005) | A breakthrough | "Old wine in a new bottle"; later approved for three cancers; $1B+ in annual sales | Mixed |
| Cancer MoonShot 2020 | Vaccine trials in 20,000 patients | STAT: "minuscule and vague findings"; GPS Cancer marketing | Overstated |
| NantKwest IPO (2015) | Largest biotech IPO by valuation | Fell more than 80%; traded near $1 | Overstated |
| Zoom (2013) | Quiet early investment | $50M valuation grew to $70B | Delivered |
| Anktiva (2024) | NK-cell activation works | 71% complete remission; FDA approval; 2026 warning letter on promotion | Delivered Caveat |
| L.A. Times (2018) | Save the institution | Six Pulitzers; ~$1B invested; two layoff rounds; endorsement revolt | Mixed |
Tactic: Promote only claims you would repeat to the harshest expert in the field, because the gap between your claim and your data is debt that every future announcement will have to pay down.
Principle 7
Answer in the other language
Witt's "Soon-Shiong uncertainty principle" (a business answer to a medical question, a medical answer to a business question) can be read as evasion. It can also be read as the competitive advantage of having more than one domain. A surgeon who ran factories could talk to oncologists about manufacturing and to bankers about mechanisms, and in each room he was the only person who spoke both. Jennifer Malin noticed that non-oncologists left his whiteboard sessions convinced he was a genius. Only specialists found parts of it "kind of wacky."
This works as long as no one in the room is fluent in both. It fails, at the FDA, with specialist reporters, before a jury, once someone is.
Tactic: Build fluency in at least two domains that rarely talk to each other, use it to translate between rooms rather than to escape scrutiny in either, and assume that someone fluent in both will eventually be in the room.
Principle 8
Proximity precedes the deal
Nearly every turn in the career starts with physical proximity: Iacocca in the VA lab, the locker room forty-five minutes after Kobe Bryant's injury, the hospital bedside of Burt Bacharach's son, courtside seats, Bedminster twelve days after an election, a podium beside Cyril Ramaphosa. Soon-Shiong bought Magic Johnson's 4.5 percent of the Lakers, which came with four courtside seats, and treated it as infrastructure for relationships rather than an investment. Friends describe him gently asking to see the chart. Critics describe him walking onto someone else's ground. Both are describing the same move.
The skill is showing up at the moment of maximum need with something useful: a diagnosis, a guarantee, a phone call to the right surgeon. It is a doctor's habit carried into business.
Tactic: Put yourself, with real expertise to offer, in the rooms where decision-makers face their most acute problems, and help before you ask for anything.
The NantWorks model was an ecosystem in which his foundation funded institutions, those institutions contracted with his companies, and his companies supplied one another. It is efficient on a whiteboard. In practice it produced the Utah audit and the "Cinderella-slipper" quote, Politico's finding that more than seventy percent of the foundation's spending went to entities he controlled, a thirty-one percent two-day drop in NantHealth's stock, and questions about Verity Health's twenty-million-dollar IT contract with a vendor in which he had a stake. The Los Angeles Times deal itself came with a patent-licensing agreement with NantWorks.
Soon-Shiong may sincerely believe, as he said of Utah, that he was "befuddled by any other motivation that could be perceived." Perception was the problem. In a vertically integrated empire, every internal transaction looks to an outsider like a conflict of interest, and sooner or later an outsider checks.
Tactic: For every transaction between entities you control, document an arm's-length benchmark and an independent approver before signing, as if a reporter will review it, because one will.
Principle 10
Your incentives will be read as your intent
Soon-Shiong's account of Cynviloq is that Sorrento's data had problems and the drug was "falling apart." Sorrento's account is "catch and kill": buy a cheaper competitor to Abraxane, let its patents lapse, and avoid $1.2 billion in milestone payments. Courts and arbitrators will decide which is true. In public, the structure of his holdings had already decided. He was a large Celgene shareholder, Abraxane cost $1,378 a dose in the U.S. against Cynviloq's $180 in Korea, and the drug never launched. When your portfolio benefits from a project failing, observers will assume you wanted it to fail.
Tactic: Before acquiring an asset that competes with something else you own, publish binding development commitments or bring in an independent co-owner, so that your incentives cannot be read as the reason the asset dies.
Principle 11
Attention is the owner's scarcest asset
The Times was twenty-third of twenty-five business lines on his fact sheet. Pearlstine went from several conversations a week to almost none after March 2020. Merida's description of their schedule, "We don't have a cadence," says it all. Soon-Shiong spent close to a billion dollars and admitted he was "not really clear" on the total. Money cannot stand in for the attention of the person who controls the institution. When attention finally arrived, in October 2024, it came as a sudden intervention after years of absence, which is the most destabilizing way an owner can engage.
Compare the factory years, when his attention was total and the results were measurable. His record improves where he is most focused and deteriorates where he is merely present.
Tactic: Rank your holdings by the attention each will actually get from you, and hand control of the lowest-ranked ones to a principal with a clear mandate rather than intervening occasionally.
Principle 12
Stay on the thesis long enough to be right
He says that by 1990 he believed the natural killer cell was key to both rejection and cancer. He grew up watching his father treat the body as its own pharmacy. Over thirty-four years, through the diabetes ban, NantKwest's collapse, the moonshot critique, and an analyst saying "No one cares," the NK-cell idea stayed at the center. In April 2024 the FDA approved Anktiva. By 2026 ImmunityBio was worth about eight billion dollars, and his stake was the main driver of a doubled fortune.
Persistence is not a strategy by itself, and the same streak has kept him attached to ideas that did not pan out. Two things distinguish the NK-cell bet. It was rooted in a mechanism he had watched at close range for decades, and he kept it funded by building other businesses that did not depend on it.
Tactic: Pick one scientific or market thesis you would still hold after ten years of being wrong, finance it from businesses that don't depend on its success, and refuse to let any single failure end it.
For readers who want the triumphalist version of this arc, the paperback hagiographies are available. PATRICK SOON-SHIONG: BOUNDLESS: From Operating Rooms to Newsrooms—The Surgeon Who Took On Cancer, Big Tech, and the Global Health Divide is representative of the genre. The court filings, the Annals of Oncology review, and the Utah audit are a more useful supplement.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIQuotes and Maxims
In their words
— Patrick Soon-Shiong, to The New Yorker (2021)To me, as a kid, everything was a circle—there's no beginning and no end. And what I mean by that, as a systems engineer, is really looking at integrating—connecting dots.
— Patrick Soon-Shiong, to CNBC (2009)That's the most exciting time in life, when people tell me I can't do that. I don't know what it is about me, when you say, "You can't do that," because then I say, "Well, that's great, because then that's exactly what I should do."
— Patrick Soon-Shiong, on his reverse mergers, to The New Yorker (2021)The reverse merger has nothing to do with money or stock. It's to do with putting the right ingredients into the right mixing bowl. So you can cure patients.
— Patrick Soon-Shiong, letter to readers of the Los Angeles Times (June 17, 2018)I believe that fake news is the cancer of our times and social media the vehicles for metastasis.
— Neal ElAttrache, orthopedic surgeon, to The New Yorker (2021)Patrick functions on the edge. You need people like that. I ask myself, "Is there some kernel of genius in there that can help the people I need to take care of?" So I listen to him. I definitely listen.
Maxims
- Own the core. Give away the periphery to finance the work, but never the patent, the drug, or the vote.
- Buy the broken factory. A distressed operating business you can fix will fund an expensive idea more patiently than any investor.
- Structure is a tool, not an identity. Merge, split, and recombine entities when it serves a measurable end, and expect to explain each move.
- Disclose before you're discovered. The folder you hand over yourself does less damage than the one a rival hands over.
- Approval can be the moat. In administered markets, a modest improvement that earns a code can outsell a better product that doesn't.
- Hype is borrowed money. Every claim larger than your data has to be repaid, with interest, when the specialists read the results.
- Translate between rooms; don't hide in them. Fluency in two fields is an advantage until someone fluent in both arrives.
- Show up at the moment of need. Access is earned in locker rooms and at bedsides, not in boardrooms.
- Related parties will be audited by the public. Benchmark every internal deal as if a reporter will read it.
- An owner's absence is a decision too. An institution you don't attend to will eventually be run badly, or abruptly, by you.
