Search the KHIT Blog

Monday, October 19, 2015

Holmes is no Sherlock. Edison is no Watson. Theranos, continued.


Continuing to follow the story. It's gonna be a tough week at 1701 Page Mill Road, Palo Alto.

From Fortune,
Ex-Apple exec describes first-hand experience using Theranos technology
by  Michal Addady

Jean-Louis Gassée, former head of Apple engineering in the late ’80s, wrote about his personal experience using the technology of Theranos, a multibillion dollar health care startup that aims to revolutionize blood testing. Last week a major controversy arose over the company’s claims.

Gassée has a condition that requires him to get his blood checked often. He was attracted to Theranos because it markets itself as an inexpensive and painless alternative to regular blood tests. After visiting the hematology lab at Stanford University, Gassée went to a Walgreens store to try out the Theranos method. The results he received were significantly different than what he saw at Stanford.

Gassée continued his experiment by returning to both locations the next day to test his blood. The Stanford results had changed, but not significantly so. The Theranos results, on the other hand, were strikingly different from both the Stanford results and the previous day’s Theranos results.

Curious about the reliability of this technology, Gassée wrote a letter to Theranos founder Elizabeth Holmes, stating his findings and saying, “I find the price and convenience of Theranos services attractive, but I worry about the reliability of the important HCT number,” referencing the “hematocrit” number, which measures the volume percentage of red blood cells in the blood. He asked about the company’s methodology, standards, and quality controls, and gave Holmes a chance to respond before publishing an account of his experience. He never received a response, he said...
Lab Test Firm Theranos Offers Weak Rebuttal To WSJ Charges
By Clayton Browne, ValueWalk


Medical lab test company Theranos offered a rebuttal to the critical article about the firm published in the Wall Street Journal last week, but most analysts and experts were not impressed.

The “rebuttal” was notably short on details or proof, and Theranos continues to claim any kind of details about its new Edison small-sample blood testing technology, including the number of tests that are currently performed with Edison, are “trade secrets”, which just adds to the growing cloud of uncertainty and potential fraud surrounding Theranos and it ultra-ambitious founder Elizabeth Holmes...
$10 Billion Health Startup's Board of Directors Lacks Medical Expertise
by Lydia Ramsey, Slate Business Insider

...The company, which is unlike any other, also has a board whose makeup is unlike that of most other health companies. Its members are:
  • George P. Shultz—former U.S. secretary of state
  • Gary Roughead—retired U.S. Navy admiral
  • William J. Perry—former U.S. secretary of defense
  • Sam Nunn—former U.S. senator who served as chairman of the Senate Armed Services Committee and the Permanent Subcommittee on Investigations
  • James N. Mattis—retired U.S. Marine Corps general
  • Richard Kovacevich—former CEO of Wells Fargo
  • Henry A. Kissinger—former U.S. secretary of state
  • William H. Frist—heart and lung transplant surgeon and former US senator
  • William H. Foege—former director of the Centers for Disease Control and Prevention
  • Riley P. Bechtel—chairman of the board of the Bechtel Group Inc., a construction company
  • Sunny Balwani—president and COO of Theranos
  • Elizabeth Holmes—CEO and chairman of the board of Theranos
To make sure we got all that: that's six former government officials, two former military leaders, two corporation leaders, two members of Theranos' leadership, and two men who graduated from medical school. As my colleague Kevin Loria writes over at Tech Insider:

"The caliber of the board suggests that Theranos must have developed a transformative innovation, but other than Frist, who has not practiced medicine in many years, only Foege is a medical professional."...
Here we go again with those gushing "Free Beer Tomorrow" words and phrases. "Transformative innovation," "revolutionize,' disruptive," etc. One of my Facebook comments over the weekend:
This is what you get from inadequate regulation and transparency. I cover a lot of this high-tech cutting-edge healthcare stuff for my blog. I know that Theranos has repeatedly rebuffed Health 2.0 requests for them to present at the H2.0 conferences.

And, as I observed on another thread: What would be truly "innovative" would be widespread availability of the so-called "lab on a chip" smartphone technology, where you could cheaply do your own blood and urine tests at home without any need for any intermediary whatsoever. The Theranos business model still sees THEM as the intermediary, simply taking market share away from the large incumbent laboratory vendors. That doesn't strike me as all that "innovative" over the long haul. And, now, if it is in fact verified that they've been gaming their PE data, then they have a very, very serious problem. And not just with the regulators. The big IPO payday may never happen. Their responses thus far are not very reassuring. It will be interesting to follow developments in the story. A big Epic Fail may be in store.
Going back to Matthew Holt's THCB post that started this thread:
Whether or not Theranos is lying, cheating, not using its own tech, or its cool stuff just doesn’t work, the trend towards comprehensive, cheap and soon at home lab testing is clear. More than 5 years ago a company called BioIQ was selling at home fingerstick based cholesterol & glucose tests. In the past year the two stage Nokia Sensing XCHALLENGE (of which we hosted stage 1 at Health 2.0 in 2013) has revealed a plethora of companies taking minute quantities of blood, pee or spit and doing complex diagnosis from them. And it’s not stopping there. The next phase is using light and other sensors to diagnose direct from the skin. Whether or not the locus of activity ends up using Theranos at Walgreens or the kitchen table using something else, the dam holding back continuous, cheap multi-faceted testing is going to burst soon. 
Yeah.

It seems to be that to be honestly classified as an "innovative, transformative disruption," a health care technology would have to be [1] less expensive, [2] more convenient for the customer/patient, and, [3] more accurate.

[1] Theranos touts their pricing transparency and "lower" costs.
"You wouldn't go to a grocery store that didn't have price tags. Why should getting a blood test be any different? At Theranos our prices are always online and available. They are a fraction of other labs. And if you have insurance they are often less than the co-pay. It's all part of making lab testing as accessible as possible."
Fine. I note that they list their PSA assay at $12.65. LabCorp notified me last time they'd be charging $198 for that single parameter test. Notwithstanding that I had to leave a credit card number with them, they subsequently accepted about $19 contract rate from BCBS/RI (no balance billing). So, yeah, Theranos would be about a third lower.

[2] Convenience. While not fully "DTC," (Direct to Consumer), if all I have to do is go to a Walgreens rather than shlepping through another several-hours-out-of-my-day primary care visit just to get a lab slip, that's clearly more convenient (and much cheaper, net, particularly given my not having to incur another 99213 encounter charge, something I griped about in my "Shards" post).

The other "convenience" facet touted by Theranos is the putative quicker turnaround time. Well, I now have to submit to an annual "wellness screening" encounter at my wife's office so we continue to get our BCBS/RI premium "discount." Yes, it's more "convenient" (and cheaper) than going to the doc. They sloppily do your Vitals (BP through my right arm shirt sleeve? Seriously?), allow you to self-report height and weight (seriously?), and then whip out a thin sterile glass pipette (about the size of a cocktail swizzle stick) with a sharpened beveled business end. A quick alcohol swap on a fingertip, followed by a quick little stab, and you watch the blood flow into the pipette, which is subsequently inserted into a small desktop device about the size of a cable set-top box. A couple of minutes later you have cholesterol parameter estimates.

By no means an Edison analytic Swiss Army knife, but, same "convenience" idea. "Revolutionary"?

[3] Accuracy. That's really the entire regulatory rub here. The rest is just marketing copy/venture capital puffery.

How do we determine whether bioassay specimen results are tolerably "accurate"? In particular those now being churned out by the now-beleaguered Theranos?

Well, there's a mature QA/QC science going to that, regarding which I wrote in my 1998 grad thesis. e.g.,
The terms “accuracy” and “precision” are not synonyms. The former refers to closeness of agreement with agreed-upon reference standards, while the latter has to do with the extent of variability in repeated measurements. One can be quite precise, and quite precisely wrong. Precision, in a sense, is a necessary but insufficient prerequisite for the demonstration of “accuracy.” Do you hit the “bull’s eye” red center of the target all the time, or are your shots scattered all over? Are they tightly clustered lower left (high precision, poor accuracy), or widely scattered lower left (poor precision, poor accuracy). In an analytical laboratory, the “accuracy” of production results cannot be directly determined; it is necessarily inferred from the results of quality control (“QC”) data. If the lab does not keep ongoing, meticulous (and expensive) QC records of the performance histories of all instruments and operators, determination of accuracy and precision is not possible...

A “spike” is a sample containing a “known” concentration of an analyte derived from an “NIST-traceable” reference source of established and optimal purity (NIST is the National Institute of Standards and Technology, official source of all U.S. measurement reference standards). A “matrix blank” is an actual sample specimen “known” to not contain any target analytes. Such quality control samples should be run through the lab production process “blind,” i.e., posing as a normal client specimens. Blind testing is the preferred method of quality control assessment, simple in principle but difficult to administer in practice, as lab managers and technicians are usually adept at sniffing out inadequately concealed blinds, which subsequently receive special scrutiny. This is particularly true at certification or contract award time; staffs are typically put on “red alert” when Performance Evaluation samples are certain to arrive in advance of license approvals or contract competitions. Such costly vigilance may be difficult to maintain once the license is on the wall and the contracts signed and filed away...
I just would like to see Theranos' PE specimen and internal QC data. How about some "transparency" there?

Why should I not be holding my breath?
Is Theranos $9 Billion Worth of Hype?
by Steve Tobak, FoxBusiness

"[T]he company has never published any peer-reviewed papers in medical journals discussing its methods or demonstrating how its results compare with traditional methods."
I hope they can satisfactorily explain all of this. I'll repeat: If Theranos gamed their PE samples and internal QC data and the CEO knew it, they have a very big problem.
__

UPDATE
FDA tells Theranos to stop using its tech except for approved herpes test
By Nicole Gray | October 19, 2015

  • The FDA recently showed up at Theranos, which is based in Palo Alto, CA, and told the company that the "nanotainers" the company uses to collect finger-pricks of blood are considered unapproved medical devices.
  • Theranos has submitted more than 100 proprietary tests to the FDA for approval, but only one—for herpes— approved.
  • Adding to the company's woes, Theranos has also been audited by the Centers for Medicare and Medicaid Services (CMS), which is the main regulatory overseer of clinical labs...
Stay tuned.

UPDATE, WBUR

 
"Eventually there is no place to hide." - Michael Moritz

From the San Jose Mercury News' Michelle Quinn:
Quinn: Time for Theranos and Elizabeth Holmes to open up
By Michelle Quinn


Her firm, Theranos, promised to revolutionize consumer health, starting first with making blood testing easier and more affordable. But right now, its name, a mixture of the words "therapy" and "diagnosis," and its lofty promises are in crisis mode and it will take more than good marketing and lawyering to steady the ship...

"Holmes' drive to change the world has a familiar ring here," I wrote at the time. "She may actually do it -- or not. It's always hard to know in Silicon Valley whether the hype matches the reality."

And now, doubts are reaching a fevered pitch with critics and tech pundits questioning the fundamentals of Theranos as well as its prospects for growth...
Theranos' integrity matters to the people who rely on the company for testing of their blood -- and all the personal diagnostic information contained therein, of course.

But there is another reason for its integrity to matter: Investors, students and entrepreneurs in the diagnostic testing realm and beyond need to know what is real about Theranos and Holmes, which have received a considerable amount of attention...
Excellent piece. Read all of it.

No one's asking them to show their schematics and engineering drawings to the regulators and competitors. Just performance evaluation specimens with a tight chain of custody, so that we can be sure the samples were actually run on their own (or industry-standard) equipment. This whole "proprietary" secrecy thing is a red herring. 

Cheryl and I both worked for a number of years in a laboratory in Oak Ridge, where a lot of our bioassay samples were to serve as evidence in court cases (radiation and mixed waste contamination and exposure litigation). So, we know exactly how this quality control and proficiency examination process works. We analyzed water, soil, vegetation, air, milk, blood, urine, tissue, and bone matrices, and trained to a forensic standard (Frye Standard at the time). The commercial "clinical" standard this company has to follow is not nearly that strict. If they can't even hit CLIA, CMS, and FDA proficiency/QA targets, they shouldn't be in business.

As Michelle noted in her article, it's time to come clean.

JUST IN: FROM FASTCOMPANY

Update 10/21/2015 5:28 p.m. ET: The Journal has responded to Holmes' reaction to its articles with a statement:
We’d like to thank Theranos CEO Elizabeth Holmes for agreeing to an interview today at our WSJDLive 2015 conference, and for the opportunity she gave all of our readers and viewers to hear from her. We always seek full input from the subjects of stories, as we did repeatedly throughout our reporting on Theranos.

Nothing said at the conference by Ms. Holmes refutes the accuracy of the reporting done by John Carreyrou or of the articles, which were subject to the Journal’s rigorous and careful editing process. Contrary to Ms. Holmes’s claims, the Journal shared all facts and anecdotes published in the articles with Theranos before publication, in accordance with our longstanding editorial practice and principles. The company was given plenty of opportunity to respond. Ms. Holmes declined interview requests from the Journal for more than five months, but the general counsel and outside counsel of Theranos provided significant input, which was fairly reflected in the articles.

We note that Ms. Holmes sought to challenge the reliability of our sources, but it remains the fact that she doesn’t know from whom the information for our articles was gathered. We assure her and our readers that our sources were well positioned to know the information they provided about Theranos, and they were vetted before publication.

The Journal reiterates that our articles about Theranos were thoroughly reported, fair and wholly accurate.
Ouch. Wow. I'm sure the Journal is fully up to speed on things like libel, so Theranos is gonna have to do better than they've done thus far.

ERRATUM

Coming up shortly, thoughts on my current read.


apropos of several of my prior posts, e.g., here, here, and here. Also newsworthy in light of this: "Watch out, Theranos? Edico Genome CEO backing stealth one-drop-of-blood startup," and this new initiative, "DNA.land."
___

More to come...

No comments:

Post a Comment