Patrick Soon-Shiong, the California health care billionaire, believes the United States is fighting a flawed war on cancer, "stuck in dogma." His bracing critique caught the attention of Joe Biden and, more recently, Donald Trump, who met privately with Soon-Shiong twice during the transition, as he reportedly angled for a role in the administration.
Soon-Shiong believes he has found a pathway to turning the disease into a manageable condition, commanding his own multibillion-dollar network of for-profit and nonprofit groups that combines health information technology with genomic testing. He compares himself to the world's great medical philanthropists, saying he wants to emulate the highly respected Howard Hughes Medical Institute.
Story Continued Below
But Soon-Shiong's research foundation — called the Chan Soon-Shiong NantHealth Foundation, after his wife — hardly resembles the Howard Hughes or any other grant-making charity. A POLITICO investigation found that the majority of its expenditures flow to businesses and not-for-profits controlled by Soon-Shiong himself, and the majority of its grants have gone to entities that have business deals with his for-profit firms.
The deals have raised alarms among some tax specialists, who question Soon-Shiong's use of tax-free dollars to boost the bottom lines of his for-profit businesses. The deals also cast a sharp spotlight on the man who seeks to be the public face of emerging cancer treatments and a major player in Washington's battles over federal research spending.
The POLITICO investigation found that:
• Of the nearly $59.6 million in foundation expenditures between its founding in 2010 and 2015, the most recent year for which records are available, over 70 percent have gone to Soon-Shiong-affiliated not-for-profits and for-profits, along with entities that do business with his for-profit firms.
• The foundation sold a prime office building to Soon-Shiong's for-profit firm for $6 million, just six years after it had been sold for $13.3 million and later deeded to the foundation, according to tax records. At the time of the $6 million sale, it was assessed for tax purposes at $10.7 million.
• Six employees of Soon-Shiong's for-profit companies were also paid by the foundation, which raises questions of whether the foundation is covering overhead for his for-profit firms, according to tax specialists.
• The foundation contributed $3 million out of a total of $12 million donated by Soon-Shiong-controlled entities to a University of Utah program to map the genomes of 1,000 state residents. University officials say they let Soon-Shiong's entities write the grant specifications. The specifications gave a major advantage to his for-profit firms, which got the $10 million gene-mapping contract.
• Soon-Shiong-controlled charities gave a total of $15 million — including $10 million from the NantHealth Foundation — to a fund that benefited Phoenix Children's Hospital, which concluded a pair of deals with Soon-Shiong's for-profit companies for many millions of dollars.
In responses to POLITICO's questions, Soon-Shiong defended his record of charitable giving and stated that many of his donations provided no tax benefit to himself. His not-for-profits have complied with tax law, he said.
"All of these endeavors, including my charitable endeavors, are primarily done so with the hope that my efforts and my funds will work to reduce and even prevent the deadly consequences of multiple forms of cancer, as well as other debilitating and deadly conditions or diseases," Soon-Shiong said.
Yet tax experts express doubts about some of the arrangements.
"The abuse is taking money that is supposed to be irrevocably dedicated to charitable purposes ... and using it for other, self-benefiting purposes," said Lloyd Mayer, a Notre Dame law professor specializing in not-for-profit law. For some private foundations, he said, "[D]onors tend to think of the foundation's money as 'their' money, which they can use for any purpose, when in fact it is no longer their money because it can only be used for charitable purposes."
From surgeon to mogul
A South African émigré and former UCLA transplant surgeon, the 64-year-old Soon-Shiong has, in recent decades, become one of the richest men in medicine, with an estimated fortune of about $9 billion.
He has increasingly taken on a public role, currying favor with politicians of both parties. He sat on the blue ribbon panel for Biden's cancer moonshot, and touts his closeness to the former vice president, suggesting that he offered advice about the treatment of Biden's son, Beau, and inspired the term cancer moonshot.
In addition, his family foundation partnered with the Bill, Hillary and Chelsea Clinton Foundation, with Soon-Shiong sharing the stage with the former president on several occasions. Soon-Shiong and his wife and children gave substantially to Hillary Clinton's presidential campaign in 2016.
But that didn't stop Trump, in his search for out-of-the-box policy thinkers, from sitting down with Soon-Shiong at both Trump Tower and Trump's golf club in New Jersey during the transition, amid reports that Soon-Shiong was under consideration to be head of the National Institutes of Health or White House adviser. White House officials told POLITICO that Soon-Shiong is not currently a candidate for either post.
Soon-Shiong, however, isn't the typical billionaire seeking political influence to advance his ideology or help his business. He has a larger agenda — reordering research priorities. Growing up in apartheid-era South Africa, the son of Chinese immigrants, Soon-Shiong studied medicine at the University of the Witwatersrand. Graduating at age 23, he immigrated to Canada and earned a master's at the University of British Columbia. From there, he was recruited to UCLA — which is where his reputation for innovation and buzz got established.
While at UCLA, in 1993, Soon-Shiong attempted a bold approach to cure diabetes by transplanting pancreatic islet cells into a patient. The operation was temporarily successful: For a time, the patient no longer needed insulin shots. Amid the ensuing media attention, however, the American Diabetes Association scolded Soon-Shiong for "inappropriate hype." The patient later committed suicide after the effects of surgery wore off.
Soon-Shiong went on to help create a new cancer drug, Abraxane, which oncologists cited as an incremental improvement over previous drugs. It's now used for certain forms of lung and prostate cancer. The company he created to sell the drug, Abraxis, was bought by the drug giant Celgene in 2010 for an estimated $3 billion. Two years earlier, he had sold another of his companies, American Pharmaceutical Partners, a generic drug company, to the German drug company Fresenius, for an estimated $4.6 billion.
The two sales made Soon-Shiong an exceedingly wealthy man, and he started buying things: Over the past six years, he has purchased stakes in the Los Angeles Lakers and the Tronc chain of newspapers, which includes the Chicago Tribune and the Los Angeles Times.
But buying and selling flashy assets isn't his passion. His stated ambition is to remake health care and its institutions, with his own for-profits and nonprofits leading the way. Soon-Shiong has four not-for-profits. Tracking his for-profit companies is tougher: POLITICO counted 23 companies, current and former, associated with one address alone.
A particular target is health care software systems, which he has called "medical bridges to nowhere" in a "broken non-system." Through his umbrella company, NantWorks LLC, Soon-Shiong has established firms focused on software, genomic and protein tests, and cancer immunotherapy drugs, which he hopes to sew together into cancer breakthroughs.
His multifront attack on cancer got a boost with the formation of the NantHealth Foundation, which Celgene staked in 2010 with an endowment of $41 million — including a multimillion-dollar property in Culver City, California — and the promise of $75 million in future contributions tied to business and research milestones. Celgene, in exchange, received first right to purchase any drugs coming out of the foundation. It was part of a complex transaction in which the drug company and Soon-Shiong restructured multiple assets.
The foundation, however, got off to a rocky start. It claimed in its first year's tax filings to have spent a whopping $48 million on general expenditures — an amount Soon-Shiong now attributes to an accounting error.
"The inadvertent errors had no significance whatsoever, as no tax liability would have resulted from these corrections," Soon-Shiong wrote in his statement. But the errors were hardly incidental. Notre Dame's Mayer points out that tax filings allow the IRS and the public to scrutinize the operations of not-for-profits, and they should be accurate.
"Here the discrepancies were a large portion of the reported revenues and almost all of the reported expenditures," he said.
But even the amended expenditures indicated that the NantHealth Foundation wasn't going to be a traditional grant-making philanthropy, but rather an adjunct of Soon-Shiong's larger network: Over 70 percent of its expenditures went to Soon-Shiong's companies, nonprofit organizations or institutions that did business with his for-profit firms.
Building an empire
Over the next five years, the NantHealth Foundation steadily transferred assets and funds from its coffers to Soon-Shiong's for-profit firms. These included real estate, personnel and research agreements.
The foundation paid $2.34 million to Soon-Shiong's umbrella company for various services. It also made almost $1.41 million in loans to Soon-Shiong for-profit businesses, balanced by $760,597 in loans from the for-profits. In addition, the foundation gave $24.6 million to a struggling data network controlled by Soon-Shiong, which in turn paid his for-profit firm $2.1 million in management and service fees.
Marc Owens, a former head of the Internal Revenue Service's tax-exempt organizations division, said, "The evidence suggests that the medical research organization is operating for the private benefit of the related for-profits to more than an insubstantial degree."
Perhaps the largest transfer of assets from the foundation to Soon-Shiong's umbrella firm was a piece of property — 9920 Jefferson Blvd. in Culver City — that was part of Celgene's initial start-up contribution. The foundation had it appraised twice, Soon-Shiong said, at an average of $6.07 million. The foundation then sold it to Soon-Shiong's umbrella firm for that price in 2012.
There are reasons to doubt whether $6.07 million was a fair price. According to the Los Angeles County Assessor's website, the property had been sold six years earlier, in 2006, for $13.3 million, and in 2012 it was assessed at $10.74 million for property tax purposes.
Since the sale, Soon-Shiong has received hundreds of thousands of dollars in rent on the property, and at least 23 Soon-Shiong-owned companies have called it home. One of Soon-Shiong's drug companies, for example, pays an annual $600,000 "license fee" to the umbrella firm and spent $3.5 million to improve the facility, according to Securities and Exchange Commission disclosures.
The foundation also appears to have shared resources with Soon-Shiong's private companies by hiring some of their employees.
Starting in 2013, the foundation used tax-free dollars to pay at least six executives who also appear to have been employed by Soon-Shiong's for-profit companies. Evidence for the overlap ranges from LinkedIn pages to corporate filings, scholarly publications, patents and news releases that identify them as employees of Soon-Shiong's for-profit firms in years they were paid by the foundation.
Soon-Shiong denies any of the employees had dual employment, suggesting they were working for different entities at different times of the year. "The employment of all employees was carefully reviewed prior to internal employment transfers to ensure no such issues with overlapping employment ... existed," he said.
But tax experts note that nonprofit organizations aren't allowed to overpay their employees the way a for-profit might — they can't be paid more than fair compensation for services rendered. If Soon-Shiong is correct, the employees — who cumulatively received nearly $1.7 million from the foundation — received hundreds of thousands each for working only part of a year. That "raises a question about the reasonableness of the compensation paid by NantHealth Foundation," Mayer said.
Questions about tax status
Those and other factors raise questions about the foundation's tax status.
Soon-Shiong's foundation characterizes itself as a medical research organization, which allows it to have certain relationships with related entities such as his for-profit firms; without that status, it would be a private foundation, and thus barred from dealing with related entities. To qualify as a medical research organization, it must affiliate and engage in research with a hospital. It has a three-year grace period to establish such relationships.
Throughout its existence, the foundation has listed Providence Saint John's Hospital in Santa Monica, California, as its primary affiliation. While a 2011 document on file with the state describes a research agreement with Saint John's, the hospital has no copy of the document and no record of any research projects with the NantHealth Foundation, said hospital spokeswoman Lauren Lewow.
Soon-Shiong says this is legal: His foundation is required to have a relationship with a hospital like Saint John's, but "it is not necessary for the medical research organization to conduct its research at the hospital — the hospital need only be available as needed for research, which St. John's is per the agreement," he said in his statement.
The IRS' Internal Revenue Manual appears to disagree. It asserts that "there must be a joint effort on the part of the research organization and the hospital pursuant to an understanding that the two organizations will maintain continuing close cooperation in the active conduct of medical research."
Some of the factors helpful for qualifying as a medical research organization are "substantial evidence of close cooperation between staff members of the research organization and staff members of the hospital," the manual states.
Owens, the former IRS official, said, "The bottom line is that the absence of 'current or recent' joint projects with the hospital might well jeopardize the entity's status as a medical research organization."
In the 2011 California filing, the foundation also lists general affiliations with Stanford University and the University of Southern California. Spokespersons for both schools' hospitals could find no record of such a relationship, however.
Some of the filings also mention the University of California as one of the foundation's research relationships, but of the two campuses that the university's central office identified as having relationships with Soon-Shiong, a UC Riverside spokesperson said it has no records of research with the NantHealth Foundation, and a UCLA spokesperson said it couldn't provide a definitive answer.
In response, Soon-Shiong said the foundation worked with a Pennsylvania hospital called the Windber Medical Center through the Chan Soon-Shiong Institute of Molecular Medicine. However, the news release announcing the "definitive agreement" between Windber and the institute wasn't released until Aug. 14, 2015 — almost five years after the foundation began operating as a medical research organization.
If the organization is not a medical research organization, it is a private foundation, Mayer said. Private foundations have no leeway to conduct business with related for-profit organizations, and would be in violation of tax laws if they did so.
Contracts follow gifts
The NantHealth Foundation also has a pattern of making grants to entities that later do business with Soon-Shiong's companies. Since October 2010, the organization has given out $15.9 million in grants to outside institutions. Two of the grants — one to a fund set up by the city of Phoenix, the other to the University of Utah — together account for $13 million of the total. In both instances, the beneficiaries of the grants later did extensive business with Soon-Shiong's for-profit companies.
Soon-Shiong says there is nothing unseemly about charitable gifts mingling with business dealings in this case. "[I]t is only natural to assume that a Foundation or charitable entity I may control or contribute to would be involved with people or organizations that are also part of the medical world that I live in," he said in his statement.
But a review of the Phoenix and Utah relationships suggests potential ethical and legal landmines.
In September 2014, the foundation and the University of Utah signed a gift agreement for a project to sequence the genomes of 1,000 Utahans and reveal their connections to certain diseases. The project drew on $12 million in Soon-Shiong's philanthropic dollars, of which the NantHealth Foundation contributed $3 million. That amount was matched by $3 million from the Chan Soon-Shiong Family Foundation and $6 million from the National Center for Health Integration, another Soon-Shiong not-for-profit that has itself been a beneficiary of the NantHealth Foundation.
Of the $12 million, $10 million came back to Soon-Shiong-related businesses when the university chose his for-profit NantHealth and NantOmics companies to perform the genome sequencing in January 2015.
Some scientists believe the terms of the deal were not entirely driven by research needs. The signed agreement included a requirement for "18,000 throughput," technical jargon for a sequencing machine that can handle 18,000 genomes in a year. That seems excessive for a project to sequence 1,000 genomes, said Keith Robison, a pharmaceutical scientist familiar with genomics.
The requirement appeared to narrow the university's options to a company with a top-of-the-line Illumina HiSeq X10 sequencing machine, Robison said. Illumina had delivered the machines to only 18 customers by January 2015, according to Barclays analyst Jack Meehan. One of them was NantHealth.
Part of what made the HiSeq X10 so technically impressive was its low cost per gene test. However, while the HiSeq X10 can churn out tests at $2,000 per sample, Robison said, the University of Utah was paying NantHealth $10,000 per sample. Even allowing for overhead, it "seems like a premium price paid for a commodity item," he said.
"It looks like the gift agreement was structured very carefully and cleverly by the foundation to make sure the university in reality had no choice but to hire NantHealth to do the sequencing," said Notre Dame's Mayer, after reviewing the material related to the deal.
Asked whether the terms of the deal effectively shut out labs that didn't own a HiSeq X10, University of Utah spokeswoman Julie Kiefer said the university "did not write the specifications." Based on the university's expertise, she said, "It was concluded that NantHealth was the only facility capable of meeting the required specifications."
Kiefer directed further questions to the Chan Soon-Shiong Family Foundation, which she said wrote the specifications for the mapping project.
But Soon-Shiong said the Utah scientists had demanded the sequencing standards that were written into the deal. He insisted that his companies did not profit from the arrangement.
NantHealth categorized the money coming in from the deal as "deemed capital contributions," rather than revenue. NantOmics, however, recognized its cut of the deal as revenue, which accounted for "substantially all" of its 2015 revenue, according to its SEC filings.
Whatever the label put on the funds, Mayer said, the arrangement raises questions about whether the the Nanthealth Foundation was in violation of tax law, which prohibits not-for-profits from unjustly enriching related for-profits.
A pair of donations to Phoenix Children's Hospital raises similar concerns. In December 2014, the foundation donated $10 million to a fund set up by the city of Phoenix to accept private donations that would be triple-matched by the federal Centers for Medicare and Medicaid Services and passed on to the Phoenix Children's Hospital. Shortly thereafter, Soon-Shiong contributed an additional $5 million from his family charity.
The city asked donors to sign a conflict-of-interest form stating that they were not "supplier[s] of health care items or services" or acting on behalf of such suppliers. Although Soon-Shiong signed the certification, he had recently completed a large deal with the hospital — creating a pediatric cancer research consortium that is owned 50-50 by the hospital and Soon-Shiong's company.
In its most recent financial disclosure to bondholders, the hospital values its investment in the profit-making venture at roughly $30 million. A spokeswoman for the hospital says it expects its return on investment to come from "identifying and offering the best and most advanced treatments possible for each of our patients."
A few months later, the hospital announced yet another deal with Soon-Shiong, saying it would be using his companies to provide genomic sequencing for cancer care.
The contributions from Soon-Shiong's nonprofits, combined with the business deals with his for-profit firms, raised questions about whether the donations were meant to take advantage of the triple-match in Medicaid funds to reward his companies.
For his part, Soon-Shiong said he was speaking only for the foundation and his family charity in disavowing any potential conflicts of interests when he signed the forms for the city of Phoenix; while his for-profits did business with the hospital, his nonprofits did not.
Soon-Shiong's answer puzzles Rutgers University law professor David Frankford, a specialist in Medicare and Medicaid fraud.
"The fact that the Foundation certified that it is not a related entity doesn't make it so," he said. "If an entity in this sea of entities provides services like testing, then it is [a supplier] and it is not clear to me why it is not an entity related to the Foundation, thereby violating the donation rules."
The city of Phoenix, meanwhile, said it did not know of any conflicts of interest when it accepted the donations. "[W]hen the City received the forms from the nonprofit foundations in 2014 and 2015, there was no information in the public domain that suggested a conflict with the information provided," said spokesman David Urbinato.
Phoenix's response also baffled Frankford. "Do they think they have no obligation but to take the documents at face value? Seriously?"
Through a spokeswoman, Phoenix Children's chief administrative officer, David Higginson, said the donations and deals were "independent of each other."
'A laboratory-type setting'
Tax law is rigorous in seeking to prevent the use of charities to funnel money to for-profit firms controlled by the same people; tax fraud is one consequence to be feared, but so too is giving an individual a chance to falsely inflate the value of one entity at the expense of another, which could mislead investors.
Soon-Shiong has been accused of such conflicts in the past.
In 2003, two of Soon-Shiong's partners — generics drug giant Mylan Pharmaceuticals, and his brother, Terrence Soon-Shiong — sued him for alleged misappropriation of funds, according to a report in Forbes. Each had invested in Soon-Shiong's diabetes research; they accused Soon-Shiong of diverting their money and using it to pay consultants to work on his cancer research. The claim was settled by an arbitrator, who cleared Soon-Shiong, according to Forbes' account.
In 2015, a few weeks after Soon-Shiong made his initial donation to the Phoenix health care fund, a pair of whistleblowers sued Soon-Shiong and his for-profit entities, saying Medicare and Medicaid funds would be used in an "unlawful and fraudulent manner" in connection with the joint venture involving cancer care at the Phoenix children's hospital.
Then, a few months later, the owner of Bayonne Medical Center in New Jersey accused Soon-Shiong of masterminding a scheme to artificially inflate his company NantHealth's value ahead of its initial public offering. The owner claimed Soon-Shiong directed money that Bayonne Medical Center owed him to NantHealth, which in turn donated to various hospitals, which then purchased NantHealth products to boost its revenues.
Neither claim was adjudicated. The whistleblowers' suit was dismissed for jurisdictional reasons, while the Bayonne Medical Center owner's claims were also thrown out as irrelevant to the medical center's original contract dispute with Soon-Shiong.
None of the accusations have led to changes in his basic approach, as he continues to defend his sprawling network as essential to achieving his dream of defeating cancer.
"Structuring my affairs with my non-profits in the fashion that I have has not aided me financially, but rather has, in many ways, been a negative to me financially from a tax perspective," he said in his statement to POLITICO. "What it does do is allow some activities I am passionate about to be conducted in a laboratory-type setting, as opposed to a pure for-profit mode. That has always been my objective, to be sure that a pure profit motivation would not dictate or influence my life's work."
In late February, Soon-Shiong put his approach on flamboyant display as he announced what he described as a momentous new cancer vaccine project at a health-software convention in Orlando. A trim man in a dark suit with abundant gray hair and a soothing South African accent, Soon-Shiong spoke at a podium in the teeming exhibit hall.
"Everything we've done over the last 15 years was built toward the goal we're announcing today," he said at the start of an elaborate presentation of his work, which led from cellular discoveries to genome sequencing to the construction of clouds and fiber highways to amass and transmit data, and ending with a four-pronged, experimental system to treat patients.
Asked for studies to back his claims, Soon-Shiong spoke generally of "Nature, Science, New England Journal, a few journals like that." Rather than studies, he presented testimonial videos from grateful patients.
As he touted the work of all the pieces of his network leading up to his cancer vaccine project, Soon-Shiong boasted, "It took us the last 10 years to show to ourselves that each of these elements could change the course of patients' lives."
"There were a lot of skeptics," he added, "who never came to understand that what we wanted to build was something never done before."