Column

There Will Be Blood

The meteoric rise and fall of Theranos provides lessons on the need for transparency in science.

By Misha Angrist

Five lawyers walk into a newsroom.

The good news is that this is not the start of the sort of cringe-inducing joke of which I am fond. The bad news is that it really happened.

In June 2015, Silicon Valley diagnostics start-up Theranos was valued at $9 billion on paper and was the darling of the biotech media, with the exception, it seemed, of one guy. Wall Street Journal reporter John Carreyrou had tried hard to land an interview with Theranos founder and CEO Elizabeth Holmes in order to ask her some hard questions about the company’s top-secret “revolutionary” technology. Holmes, despite having a healthy appetite for publicity (especially magazine covers), never agreed to sit down with Carreyrou.

Two months later Theranos sent a representative and five attorneys to the Journal’s Manhattan offices. As reported in Vanity Fair, the phalanx of lawyers accused the paper of illegally obtaining trade secrets from Theranos. For five hours, they harangued the Journal and Carreyrou and threatened them with a lawsuit.

If you’re not familiar with the Icarus-like tale of Theranos and its captivating mix of fairy tale, allegory, and farce, here’s the SparkNotes version: Spunky and brilliant Elizabeth Holmes started Theranos in 2003 at age 19 and soon dropped out of Stanford to pursue it full time. While she was originally interested in skin-patch drug delivery systems, Holmes took her own fear of needles as the impetus for a company that promised to transform the diagnostics industry by offering cheap, painless tests from a single drop of blood. She raised hundreds of millions of dollars, recruited ex-government heavyweights like Henry Kissinger and George Shultz to serve on the board of Theranos, and filed for numerous patents. By 2013 Theranos had opened dozens of clinics inside Walgreens pharmacies, mostly in Arizona, where patients could order tests easily and for a fraction of the going price. In July 2015, Theranos received FDA clearance for the first of the more than 240 tests on its menu. All the while, the black-turtleneck-clad Holmes was hailed as “the next Steve Jobs” (a curse if ever there was one).

In late 2014 Ken Auletta wrote a mostly flattering profile of Holmes in the New Yorker. In it Holmes described how Theranos technology, uh, works: “A chemistry is performed so that a chemical reaction occurs and generates a signal from the chemical interaction with the sample, which is translated into a result, which is then reviewed by certified laboratory personnel.” Auletta described this passive-voice-drenched, abstract account as “comically vague.” In the science writing course I teach, we have a technical name for it: “bullshit.”

This was enough to raise Carreyrou’s antennae, and thus began the hard questions for Theranos and the months of investigative reporting (during which time, reported Vanity Fair, Theranos continued to threaten him with legal action). The first of his stories appeared in the Journal in October 2015, and it was a doozy: The company was relying on old technology for all but a few of its tests, in part because there were serious questions about the accuracy of its new, low-volume technology. The company’s own employees did not trust its machines, and there were allegations that Theranos had cheated on the testing required by the Centers for Medicare & Medicaid Services.

Remarkably, things got much worse. In the wake of further revelations of shoddy lab practices, CMS accused Theranos of putting patient lives at risk. The Justice Department and the Securities and Exchange Commission both launched independent criminal investigations. This past spring Icahn School of Medicine researchers published a head-to-head comparison of Theranos tests with conventional blood draws and found Theranos results wanting. The company soon voided two years of results and issued tens of thousands of corrected test reports. Forbes revised its estimate of Holmes’ net worth from $4.5 billion to zero. In June Walgreens walked away from its deal with Theranos, leading to the closure of 40 blood test centers. In July CMS banned Holmes from owning or operating a medical lab for two years (the company said it would appeal the sanctions); by then, the company was facing a flurry of civil suits. A biopic, starring Jennifer Lawrence as Holmes, was announced a few months ago; from all accounts it will be heavy on the schadenfreude.

Besides Carreyrou, Stanford’s John Ioannidis knew something was amiss. Ioannidis is a torchbearer in the crusade for scientific reproducibility and author of a seminal 2005 paper “Why Most Published Research Findings Are False.” In February 2015, eight months before Carreyrou’s first story dropped, Ioannidis wrote an editorial in JAMA in which he worried about grandiose claims of biomedical progress being made outside of the peer-reviewed literature. Exhibit A: Theranos.

“The company’s own employees did not trust its machines, and there were allegations that Theranos had cheated on the testing required by the Centers for Medicare
& Medicaid Services.”

From day one, when asked about the details of her company’s technology, Holmes’ reflex was to deflect and say that the FDA would be the final arbiter. That approach is tenuous enough for obtaining regulatory approval; for advancing knowledge and health, it is a cop-out at best. And given Theranos’ stonewalling of both FDA and CMS in 2015, it became a joke. Science relies on transparency. Without publications open to public scrutiny and with only a portfolio of generic patent applications on the books, how was one supposed to evaluate the merits of a Theranos test?

In August many thought that answers were finally forthcoming. Holmes was to appear at the American Association of Clinical Chemistry’s annual conference; at long last, she would present real data and answer questions about Edison.

Except … she wouldn’t. Holmes did not address the prior year’s worth of questions, but instead introduced a new product. The “miniLab” is heir to Edison and meant to be a small-volume instrument for running various blood tests. Holmes said, for example, that a miniLab test for the Zika virus had been submitted for FDA approval. She also said miniLab data would be submitted for peer review. Okay … one cheer?

The audience was mostly underwhelmed (or outraged that Theranos was given a forum at all). The technology Holmes showed was smaller but not really either new nor faster. Commentators described it as a “pivot,” a “Hail Mary,” “buffing a dirty lantern,” and a “bait and switch.” One bioethicist likened it to the scene in The Wizard of Oz when the “wizard” is found out and says, “Pay no attention to that man behind the curtain!”

Will miniLab be validated? Will the Theranos narrative ultimately be riches-to-rags-to-riches — the misunderstood company that changed the way we do blood tests? Or will it be a less interesting story about a mirage — the turtleneck behind the curtain — that tried to intimidate and bamboozle its critics and then change the subject?

I have no idea, but I’m looking forward to the denouement. Pass the popcorn.