Stealth Research and Theranos: Reflections and Update 1 Year Later | Hacker News Reader

Stealth Research and Theranos: Reflections and Update 1 Year Later

jama.jamanetwork.com

52 pointsmmastrac17 comments

I would love to hear more of the reasoning as to why Theranos thought Dr. Ionnadis, famous for being hugely skeptical of medical research claims [0], would be someone to reach out to (assuming they weren't intending to just buy him out). There are plenty of other medical experts who would have cooperated with Theranos (given its pedigree) before the WSJ investigation.

[0] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182327/?report=...

Holmes probably (delusionally) believed she could convince Ioannidis to recant his comments in the February 2015 JAMA article regarding "the rationale for promoting massive diagnostic testing," and maybe even the criticism of stealth research, by demonstrating how ground-breaking their technology was, and that the IP needed to be "protected."

At its time of publication, that article was (I believe) the only one in the scientific literature that said anything about Theranos, and the company seemed to be pretty disconnected from the mainstream biotech research community. Holmes was probably only vaguely aware of Ioannidis' background.

Theranos needs to prove that a blood sample oozing from an open wound (a finger prick) is in any way comparable to sterile blood extracted directly from a vein. They also need to show that analyzing a diluted and contaminated blood sample with the Siemens instrument produces an accurate result.

At this point, I think we should assume that because Theranos won't publish any studies about the scientific foundations of their work in peer-reviewed journals or preprint sites, they have nothing of value to show us.

Stealth research is a byproduct of the weakening of the patent system. It used to be that, for something like this, you got a patent before talking to VCs. This allowed more review of whether the technology actually works.

Now we have VC money being invested based on how well someone can pitch. This doesn't work for things which are technically hard. Hence Theranos, uBeam, and possibly Cruise.

I'm trying to think of a company that proves the opposite, but I can't.

For example Tesla and Space X seem like likely candidates, but Musk, and any founder with a billion dollar + exit, is an absolute cheat code in fund raising and can't be used as a standard for any normal company. And yes, most of the companies you mentioned are completely normal with founders who are categorically not Elon Musk, or in the case of Theranos, not Steve Jobs.

Has the patent system really weakend ? or it's simply that today, the options to build something are so varied, and people have more knowledge of that, so you cannot patent them all ?

At least three things made it much more expensive to enforce a patent:

1. The America Invents Act (2011), with its new post-grant review provisions. Now, if you try to enforce a patent, the infringer can tie you up for a few years with review proceedings, which means the inventor has to spend money to defend the patent.

2. The eBay Inc. v. MercExchange, L.L.C. decision (2006), which made patent infringement injunctions much harder to get. This decision basically means that the worst thing that can happen to an infringer is that someday they have to pay royalties. It's essentially compulsory licensing of patents. So infringers have no incentive to negotiate.

3. The In re Seagate Tech decision (2007) and some related decisions changed the standard for willful infringement (and triple damages) to include a requirement of "reckless disregard", which is almost impossible to prove. These decisions also reduced the power of juries in patent cases.

It's also harder to get discovery in patent cases now, which means that if infringement can be kept a secret, it's usually possible to get away with it.

"The problem that some patients do not have tests performed when they genuinely need them is also real but probably of lesser magnitude." Anecdotally speaking, I have seen many instances when testing was procrastinated only to reveal a chronic condition (hypothyroidism, pre-diabetic sugar levels) needing immediate attention sometimes with medications.

I am genetically disposed to hyperlipidemia and like to watch my cholesterol level as a tangible indicator/reward for my life style changes.

"Better financing and organization of health care and, perhaps, reduction of the profit margin could markedly decrease testing cost, even if very old (but appropriately validated) diagnostic technologies are used."

Reduced pricing was a hype created by Theranos. When we started working in this area, we found that if one is willing to pay out of pocket some labs will offer pricing very close to Theranos.

> Theranos does stand for well thought-out and useful therapy and diagnosis and does not represent the harms suggested by another similar Greek word, thanatos (death).

This is gold. Yesterday Theranos just voided its two years of testing results by its Edison.

I saw another wordplay in that part: they say Theranos comes from "therapy" and "[diag]nos[is]". But if you combine the "ther" and the "agnos" as Greek, you get "beast without knowledge". :-O

I maintain that as this story continues to unfold, it will contain a serious criminal element at its core.

Are you an expert?

Did I claim to be?

While you make no explicit claim, I suppose the responder was reacting to your opener of "I maintain".

That implies you have been saying this all along, in a forum that we would expect to have heard it, but yet you have not been heeded.

In my perspective, there is an element to your statement that is claiming authority, thus making you appear as a self-proclaimed expert.

This is overinterpretation.

Also, I agree that it seems likely Ms. Holmes will face criminal charges. It would have been one thing if Theranos simply substituted venous blood draws for nanodraws, pivoting to not use Edison. However, they not only did that, they denied this publicly, promised to release data showing the validity of their platform (neither of this are lawsuit-worthy) and then proceeded to provide false test results at a very high rate. The problem here is that Ms. Holmes would have to have known of this- she's the CEO of the company, in very tight contact with the labs, etc, holds nearly all the power, and is the ultimate decider. Because of that, she is the direct target of lawsuits.

I expect both government action (which is already ongoing; it's likely she will be sanctioned) and class action lawsuit.

Obviously, the above is speculation, based on my intuition of watching biotechs for 20+ years.

Don't read so much into so little.

Funny thing about JAMA is that almost all (?) the articles they publish don't include code nor data. The research they publish is stealth!

Reply on news.ycombinator.com