The FDA's notes from its visit to Theranos' labs don't look good | Hacker News Reader

The FDA's notes from its visit to Theranos' labs don't look good

businessinsider.com

118 pointsbpolania97 comments

The most surprising thing about this whole saga is the way two VCs openly spoke out against Theranos. Bill Maris from GVentures and Moritz from Sequoia. I can't remember the last time I saw a VC trash someone else's investment in public. So why now?

I'll speculate that there's something so distasteful about Theranos --the wild mismatch between valuation, amount raised, and deliverables-- that these two investors felt that it threatened the ecosystem itself. Like they needed to distance Sand Hill Rd from this stinker before it taints the other (still unproven) unicorns.

Or maybe it was something else, dunno. Anyone else notice it?

I suspect this is going to turn out to be a huge spectacle. One of the downsides of so much VC cash available is that it's also a lot easier to build up a very expansive sham company which doesn't have much at its core.

Basically its going to be a big spotlight on how the VC community is a gated community who are all about playing the insider game.

I'm curious: What would be the benefit of doing this? Getting investment from VC's with the hope that it can be turned into something?

The VC game in Silicon Valley isn't about investment, it's about playing the lotto. You dump money into as many companies that have even a small degree of promise, you blow them up to the point where you can find out whether or not they have market traction, and then you collect your winnings from the tiny fraction that hit it big. Not only is it vastly disadvantageous to not have your fingers in as many pies as possible but the VCs with the most money are precisely the ones who have been playing this game already and hit it big with google, facebook, instagram, etc.

Investing in 9 sham companies that fold and 1 unicorn that pays out at 10:1 or 100:1 is more lucrative than investing in 10 companies with modest, but reliable RoI's. Peter Thiel's Facebook investment paid off at 2000:1, Accel's facebook investment paid off at 800:1, Andy Bechtolsheim's google investment paid off at 17000:1, Andreessen Horowitz made 300:1 off of instagram. If you've got a good enough pitch you can get as much money as you want.

This makes sense for small investments in companies with valuation < $10 MM but how does one just throw paychecks at valuation ~10B with no significant introspection?

Theranos' Series C rounds came in at under $30 mil across three different investors. That's chump change in the VC world, if that's the price of a lotto ticket, so be it. If you're a VC who's won the lotto before you likely have over a billion in capital. Pumping $30 mil into 15 ventures doesn't even use up half of it, and if any one of those end up making true on a huge valuation while the others just wander into the desert and die, you'll have made back your lotto expenditures and more.

Looks like everyone learned their sweet lessons from the dotcom bubble...

Pump and dump before it goes bang.

What is alarming and missing from the BI story is that there were two documents issued by the FDA - one which covers the issues mentioned (more of pre-market and conduct) the second one listed here: http://www.fda.gov/ucm/groups/fdagov-public/@fdagov-afda-org... is equally damming.

This one deals with the requisite post-market aspects (surveillance, complaint remediation, etc).

For a Medical Tech. start-up to fail on post-market surveillance is in-excusable. Move Fast & Break Things should not apply to people's health and medications.

There is something very wrong when a company with that much cash fails BOTH in pre & post market situations.

Absolutely agree. Any established CRO or pharma treats surveillance and [especially] post-marketing product safety & adverse event reporting as mission critical activities. The kind where, if you don't do it properly and follow the rules, you get shut down.

(My wife has worked in this area for the past ˜15 years, in both clinical & post-marketing environments. It's serious business.)

For those of us who are not doctors, but play one on the internet: https://en.m.wikipedia.org/wiki/Postmarketing_surveillance

A large percentage of companies don't do their post-market surveillance, and the FDA has never even once revoked approval because of it. The whole FDA approval process is basically a scam.

Uhhh... that's not correct at all.[1]

Are you confusing it with post-marketing commitments to run additional clinical trials? That's not the same thing as post-marketing surveillance.

[1]http://www.fiercepharma.com/story/roche-overlooks-80000-adve...

"Not all Phase IV studies are post-marketing surveillance (PMS) studies but every PMS study is a phase IV study."

"Under the current system, the FDA has the authority to revoke approval of some drugs for which Phase IV commitments have not been honored. This authority has never been invoked."

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3148611/

http://www.citizen.org/hrg1520

The Phases apply to drugs, not devices.

The lack of factual scrutiny for such potentially life-changing technology is a shame on both ends of this story.

I mean there are 2 Theranos centers smack in the middle of SV (Palo Alto and Redwood City) - the WSJ's exposé centered largely on actually trying the tech and finding others with disparate results - where were the SV / SF / Oak proper journalists on either end of this story actually scrutinizing a $10b monster?

"Watching the media build up and tear down Theranos feels the same with the sign bit flipped. Little data, and lots of hyperbole, both times."

c/o Sam Altman's twitter and he's spot on.

(1) https://twitter.com/sama/status/655492875073884160

where were the SV / SF / Oak proper journalists on either end of this story actually scrutinizing a $10b monster?

Most journalists are not capable of narrating how e.g. a Rails application works. They're equally not capable of walking an interlocutor through e.g. Phase N clinical trials. They rely on people who actually have domain expertise to tell them what the ground state of the world is like. One would hope that a good journalist goes around and checks those representations, but this is not always true.

Every interaction I have ever had with a journalist attempting to use me as a source has decreased my confidence in the profession. I'm on like six organizations' internal lists as "Speaks English and can answer any Japan question." (My qualifications for this: I speak English, have a professional background which suggests bare minimum competence in a field which was once newsworthy regarding Japan, and then was in the NYT in that capacity. "Close enough!")

They've asked me to provide my opinion of everything from e.g. Abe's economic policies to whether the man on the street thought that the Japanese women's soccer team was counteracting the national malaise after the Tokoku earthquakes. When I give them the honest answer, which is that I'm a small software entrepreneur who does not know anything useful about nation-level economics policies and is sufficiently politically apathetic that me knowing that Abe is prime minister actually means that I must be paying more attention than usual in 2015, that does not result in them stopping the interview or taking me off their list.

Journalism is a high-prestige field which, to re-use Michael Creighton's observation, we understand is virtually never right about even the simplest consequential facts about things we actually understand, but we trust it to be accurate with regards to consequential facts about other fields, and this trust endures through every report that we have about journalists being catastrophically wrong. The alternative to trusting journalism is to believe that the New York Times and BuzzFeed hire the substantially same people to do substantially the same work for substantially the same reasons, and that the NYT is prestigious and BuzzFeed is not because the NYT has been BuzzFeeding for centuries.

There is some threshold of evidence that causes us to accept the BuzzFeed hypothesis, right? Right? God help me, I know I sound crazy, but I think I'm there.

> that does not result in them stopping the interview or taking me off their list.

Why would they take you off their list if you gave them an honest answer (I do not know)?

You clearly knew were to stop, so you did not misinform them . That is a valuable trait for an expert.

If it really is that bad, why read newspapers then? But, if you never read newspapers of any kind, how would you have any idea about what happens around the world?

There is some threshold of evidence that causes us to accept the BuzzFeed hypothesis, right? Right? God help me, I know I sound crazy, but I think I'm there.

Welcome to the ranks of conspiracy theorists. Anyone willing to admit difficult truths eventually winds up here.

Is there a time in the history of journalism where journalists were ever right about even the simplest consequential facts? That would at least explain why they're still considered a high-prestige field. Anyway, I'd love to find out how such a worthless field earned and continues to maintain it's prestige.

It seems to me that there is a trend that many people start to lose trust in the credibility of journalism. State-funded troll factories[1] probably helps with fueling this distrust. There are of course many valid reasons to criticize media publishers, but I sure hope we could avoid throwing the baby out with the bathwater.

[1] http://www.nytimes.com/2015/06/07/magazine/the-agency.html

This is the point: if there ever was a baby in that bath, it's dead now. Read what Patio11 wrote again and realize his situation isn't an isolated incident. It's the norm.

Tell me, cpach, why did you feel the need to defend the news establishment? You obviously think there's some value they're bringing to the table. What is that value?

I think parts of what the news establishment does are worth defending. Some parts are definitely despicable. I do not stand wholeheartedly behind them.

But as I said in another comment, how else would you have any idea about what happens around the world? Do you see no value whatsoever in publications such as The New York Times, The Washington Post or The Economist? I'm not talking about The Sun, Fox News or Russia Today.

I see a value in organizations that have the funds to pay for investigative journalism. A recent example is the private jet scandal within the Swedish manufacturing company SCA.[1] The newspaper Svenska Dagbladet ran a series of articles about how the upper management at SCA misused the company's private jets, costing the company loads of money. Unsurprisingly the shareholders were not happy and many key persons had to resign because of this.

Had it not been for the articles in Svenska Dagbladet, how would the shareholders ever find out about this misuse of the company's resources?

[1] http://www.thelocal.se/20150123/one-of-swedens-most-powerful...

> how else would you have any idea about what happens around the world?

Since there's nothing I can do about anything anywhere else in the world, why should I care? Furthermore, if very little the news establishment publishes is correct, how can you say you have any idea what's actually happening around the world? You don't. With the readiness with which the news establishment will publish absolute junk, what makes you think they won't publish what they're told by their corporate owners or the government? When nobody does any fact checking, it leaves them (and you) vulnerable to manipulation.

> Do you see no value whatsoever in publications such as The New York Times, The Washington Post or The Economist?

None.

> Had it not been for the articles in Svenska Dagbladet, how would the shareholders ever find out about this misuse of the company's resources?

Wow, you had to go all the way to Sweden to dig up an example for this? We're talking about the American news establishment. I have no comment on or exposure to European news. In the American media, investigative journalism is a dead art. The disclosures we see in the news are mostly done because we have a (more or less) open market which gives an incentive for all sorts of people and organizations to profit from these sorts of disclosures. That part of the system would work just as well without journalists.

> Wow, you had to go all the way to Sweden to dig up an example for this? We're talking about the American news establishment.

All the way? Has it occurred to you that not every participant in this forum is a US citizen?

He's not a fan of two's complement, I guess.

I don't have much experience with FDA, but it looks like they are arguing with technicalities. They are not saying that product is not working, just that it isn't approved by them (for this purpose, if Theranos website is to be believed [0] - "The Nanotainer™ and our devices have already been cleared by FDA for use with our HSV-1 test").

It looks like this could be solved very quickly if all parties were interested. I am quite sure that Theranos has made some powerful enemies, many of whom probably have some influence within FDA...

[0] https://www.theranos.com/news/posts/statement-from-theranos-...

I assuming hsv-1 was approved and successful because it's a pcr test requiring very little blood already. Most of the other tests require much more volume. It sounds like these tests were the ones that were supposed to be revolutionary but were not successful.

Yeah, a PCR test can use a tiny sample, and and with a modicum of care it's not going to be subject to cross contamination in taking the sample (you do have to watch out for back-end contamination).

But I can see how things like lipid panels would have contamination problems with finger-stick draws, although they say they use a combination of those and "venous blood draws using smaller needles and smaller tubes", which doesn't strike me as a big selling point. In my experience, pain is associated with not hitting the vein right the first time, and once you hit the vein, you might as well draw more than enough for the planned tests and any followup tests ordered once the first results come in.

> although they say they use a combination of those and "venous blood draws using smaller needles and smaller tubes"

That's new, they used to claim they could do it all from their Nanotainer sample. It was and still is a crock, but that's entirely what Theranos' claim to fame is.

Yikes. And true per this cited on Wikipedia's intro page: http://www.wsj.com/news/articles/SB1000142412788732412300457... Here's what they're saying right now on https://www.theranos.com/our-lab

Our tests require less blood and less pain, regardless of whether we use our proprietary finger-stick technology or a venous draw.

We do venous blood draws using smaller needles and smaller tubes.

Now, I will say that when Quest does a venous draw they use a honking big tube ... well, the normal size since they stopped using syringes (you young whippersnappers are lucky, let me tell you!), and draw well more than a mL, but for the latter, who cares? Especially if that allows them to do followup tests without requiring another draw, as happened with me a couple of months ago. And for that matter, the larger the needle, the shorter the time is has to be in you to get a given sized sample.

>It appears as if this isn't the first time Theranos has butted heads with FDA inspectors: "This document was provided as a follow-up to the failures in your design validation conducted in 2014," the FDA report says.

$10 billion valuation, and this is the best they can do?

It seems like silly season is going to end soon. This time there are a lot of solid businesses in the valley with solid revenue and solving real problems. Then there are the dilettantes and the scammers. All will be revealed as it becomes more difficult to raise money.

They may still actually have an real innovation (we need more evidence), but I do agree that this is a massive failure for such a valuable company. After the huge setback that 23AndMe suffered from their run-in with the FDA, every well-funded company under their jurisdiction should be triple-checking everything related to FDA compliance, since it poses such a grave, existential risk.

Interestingly, both 23andMe and Theranos just sat on the documents issued to them by the FDA, see https://news.ycombinator.com/item?id=10462936

Do both have the same guy working on FDA-approval?

It's the Uber aggressive approach - break the rules, try to get big faster than they can slow you down - applied to something directly ruled by a federal bureaucracy. That'll never turn out well. With the FDA you have to play the game the way they want you to or they'll instantly nuke your operation. The Uber approach only works if you're dealing with lots of little bureaucracies.

While current interest rates are below zero effectively - the raising money will be as easy.

Observation 5: "Documents were not reviewed and not approved by designated individual(s) prior to issuance."

this one is the most damning. FDA gave them the opportunity to correct this, and they sat on it for a year.

The culture of move fast and break things is fine for consumer products, but the cavalier attitude with people's lives is overstepping.

I'm tempted to argue that the SV model of tech startups essentially prohibits the creation of any functional biotech or medical device company. The compliance and process parts require real industry experience senior expertise, and time. like you said, a cavalier attitude is completely inappropriate and will never pass scrutiny.

I worked for a medical device manufacturer for about 15 years. 21 CFR 11, and the test & quality tracking requirements, make it extremely hard to pivot quickly.

Pivoting with same indications can be done, but I agree, if you have a Class II or god forbid Class III device, and you change your indications, you've got to redo much of your 21CFR11 documentation. I built a Class I device that eventually became a Class II device, and it was a tough slog to get the QSR and CFR docs in order, however, we were a team of 15, with not nearly $400m in funding.

There is an air on incompetence about this whole thing, and Holmes isn't doing herself any favors battling with poor evidence against WSJ in public.

Pivoting doesn't make a lot of sense in this space.

However it is entirely possible to be agile and (relatively) fast in this space - not by breaking rules but by designing the process to be light where it can, detailed where it must. And then paying attention to how it is working and improving it constantly. You can probably cut time to market in half relative to industry averages this way, which is nothing to sneeze at.

Pivoting after losing your reputation for quality and playing straight with the FDA makes no sense whatsoever to me. In fact, the FDA is known for getting very vindictive after that sort of thing happens.

I count this as different than the 23AndMe situation where they're arguing with the FDA over how the results can interpreted and passed on to the consumer, that's a difference in opinion that doesn't rest on the quality of the testing science.

   that's a difference in opinion

Indications for use aren't a matter of opinion. 23AndMe clearly screwed up there, but I agree it was a different nature of screw up.

You can still develop the tech without those experts. However, if you ever want to apply it to humans, you got to follow the rules.

Certainly the pivot-constantly, push-code-daily mentality is not translatable to medical technology. That said, biotech & device companies could stand to be a hell of a lot more agile. Compliance is necessary but it isn't a complete impediment to moving quickly.

Nothing drastic in such observations, unless you have never seen FDA reports. The FDA usually issues a lot of observations and calls for corrective actions - this happens even with large and experienced pharma companies.

I disagree. Device/diagnostic companies get 483's all the time. What's telling is the scope of these observations. Theranos is failing on most counts of the quality system regulations. Judging by the observations they don't even have a corrective action (CAPA) process. That's a huge red flag.

> Judging by the observations they don't even have a corrective action (CAPA) process. That's a huge red flag.

Well if they want to hire folks out there there's a huge amount of capable people ready to be poached from other companies.

I can't imagine many people being willing to move to Theranos today, as long as their current job isn't totally awful. Reputation is very important in this field; heck, if I saw a resume with someone coming on board at this period, I'd be asking them some hard questions about why they did it.

http://www.slate.com/articles/business/moneybox/2014/07/empl...

I would theorize there's no shortage of biochemists or lab techs. The graphs seem to show a gradual increase in unemployment and a gradual decrease in employment for life sciences PHD grads. 2010 was a crossover year where afterwards less than 1 in 5 grads will get a job but more than 1 in 5 will be unemployed. This doesn't count underemployment; a biochem PHD working as a waitress is technically employed.

If only the upper 20th percentile of the grads currently get jobs, lowering one companies standards to accept the 21st percentile of grads is still pretty far along the right side of the bell curve... they'll do fine. Regardless of the details, its hard to describe a market where 80% don't have industry jobs as being limited by the labor pool.

The whispers of Theranos being a shitty company, at least in the healthcare/biotech community, have been around for almost a year. To get the type of talent they need, they will have to pay a pretty big premium, but it does seem they've got nothing but cash to burn.

I'm sceptical generally of Theranos, but none of this feels in the least bit like a smoking gun. Wouldn't you expect feedback like that for a truly revolutionary product? There's no claim of fraud, there's no claim their technology doesn't work, just complaints about technicalities. Am I wrong?

The government doesn't give you "feedback." Also, out of the five observations, only the last one is a "technicality." The process documentation and design specification and validation issues are not optional. Not just in medicine, but in any engineering field where lives are on the line.

E.g.

> Observation 1: "Design validation did not ensure the device conforms to defined user needs and intended uses."

You don't build a bridge without defining the maximum carrying load you intend to use it for, or validating that it will carry that load. Getting lucky and having the bridge not fall down (at first) doesn't ameliorate that process defect.

P.S. I haven't been following the Theranos story closely. I am taking the fine article at face value.

    > The government doesn't give you "feedback."

http://www.fda.gov/ICECI/Inspections/ucm256377.htm

    > when an investigator(s) has observed any conditions
    > that in their judgement may constitute violations

Sure sounds like feedback to me.

    > The process documentation and design specification
    > and validation issues are not optional

No, but they can be iterated on.

    > I am taking the fine article at face value

The "fine article" presents small snippets of a heavily redacted article. There's not that much value to take...

I hope the industry doesn't draw the wrong lessons from Theranos. Compliance isn't hard--the expertise is out there. But the process must be followed and that might mean moving more slowly. Which is okay because your competitors have to move more slowly too.

What's interesting about Theranos is the hints that the technology may not be ready for prime time yet either. In which case moving more slowly, and taking that into account in the funding stages, is not just a reality of compliance but a reality of engineering. This stuff is hard, MVPs aren't enough, and successful ventures are going to take longer to yield a salable product than in other fields.

Gushing press, create-a-celebrity CEO, A-list "1%-er" investors... Then, questionable product/technology? Time to grab the popcorn, this should be fun!

It would be amusing, except I'm worried the dodgy tech has ruined lives.

Holmes and her crew are scanners and fabricators - and while I'd love to say they'll get their comeuppance, they're now wealthy enough to buy their way out of anything.

It's junk science, has been since day zero. Her father's USAID connections make me think there's more going on here, and that they failing could embarass elements in government.

Also the fact that the CEO/"founder" hasonly one year of college as relevant experience. She appears to be a pop star persona: a face to send to interviews to look cool in PR, fronting for the business machine in the back. And of course there was never any tech, just hopes they ride a hype was into winning Quest's contracts.

Can someone point out to me what exactly it is that Theranos does?

My understanding was that this company administers innovative ultra-small volume blood testing. Now they say that only works for the HSV-1 test (that nobody wants, and by some accounts already existed), and for the rest of the 240 tests they use normal lab procedure.

What is the great innovation here? What is it that this company has been doing for 10 years? Why is it worth $10B?

It's not just these small vials of blood. They also make their own "revolutionary" testing equipment, which they used for dozens of tests. Except it turns out their equipment was giving quite badly inaccurate results, so they quietly switched to conventional machines. Ditto for the small blood vials.

So it's a flop, basically? That's a pity. They should just come clean; people will respect them for it.

It would be the honest thing to do, but I doubt it's the ideal for investors.

How do you even get funded without a process that is fully tested? Was this a failure of due diligence? I don't understand.

How do you even get funded without a process that is fully tested?

A fully tested process? Surely you jest. Hell, you don't even need to have an idea to get funded[1].

[1]http://old.ycombinator.com/noidea.html

Since this is no longer an option I assume that experiment didn't go well. I wonder what ideas came out of that group, a post mortem would be interesting.

This would definitely be interesting. There's a UK accelerator, Entrepreneur First [0] which takes people pre-team and pre-idea. I'm not sure if they have any exits yet, but YC occasionally fund their alumni.

[0] http://www.joinef.com

Lets try it - I come with you with proof of concept nuclear fusion device that is just below break even energywise and claim that with proper engineering I can get it to 20 EROEI. Will you fund me?

It was enough for her to have some innovation that could plausibly work.

Didn't anyone (really) check before putting money at a $10 Billion valuation??

While I was still at HP, they bought a company called Autonomy for 11B USD.[1][2] No one ever knew what they did. HP wrote off 9B USD in 2012.

1. https://en.wikipedia.org/wiki/HP_Autonomy

2. http://www.marketwatch.com/story/the-mysterious-case-of-hewl...

Autonomy did a lot of hand waving ...

but at the end, they did a decent job specializing in electronic records management & e-discovery / litigation holds. Not nearly enough to justify the ludicrous price, but at the time HP was desperate for something to help them compete in any sort of "big data" market.

I also question HP's decision to announce their entry in the cloud market via the Onion [1]

1. http://www.theonion.com/video/hp-offers-that-cloud-thing-eve...

I don't understand the intricacies of FDA approval so I might be confusing things here, but why is Theranos' front page currently claiming "Theranos Receives FDA Clearance"?

1) Can we have a document from other visits to similar labs for comparison.

2) What does the $10B valuation people keep referring to has to do with anything? If they do not meet requirements should they be valued less? This does not make sense. If you do not meet standards for xyz, you shouldn't be allowed to perfom xyz. The feedback should not be around the valuation.

3) The same press that are knocking Theranos are the same press that were calling it disruptive. If anyone is to be blamed it should be them, and the government agencies that allowed testing.

For discussion sake I’m going to throw this out - The $10B valuation is the ONLY issue here.

We know that for all the test except one, Theranos uses usual diagnostic devices (from Siemens etc.) like any other diagnostic lab. So it will be something like ~99.9% of their tests are done on traditional devices. So we can safely assume that those tests are as reliable as those from other labs.

So if they are same as any other legacy industry for 99.9% of the time, then why such a big valuation?

They should be valued same as their nearest rival (which I’m sure is a giant compared to Theranos), and if adjusted for the fact that whether Theranos is going to upset them without any secret sauce (and with fewer resources) then the valuation should be near to $0.

Well, the troubling thread here is that that may not be posssible using the amount of blood Theranos collects. Then the question becomes not only why is their valuation so high (which the public doesn't need to care too much about) but "how well do these traditional tests work with Theranos samples?"

That the answer could be "not very well" (and we haven't heard otherwise) is what's most troubling to me.

Well, the troubling thread here is that that may not be posssible using the amount of blood Theranos collects.

It's say it's worse. I followed up on the general impression they were proposing to do this all with fingersticks, e.g. this citation from Wikipedia's introductory paragraph published in 2013: http://www.wsj.com/news/articles/SB1000142412788732412300457...

Based on my general knowledge of biology and medicine, I'd expect there would be a whole bunch of things you can't test that way due to general contamination from all the normal cells and their membranes that get punctured in a finger stick. Lipid panels, for example. In fact, the large venous draws normally taken, at least for the first tube, could be in part to swamp the contamination that comes from getting the needle into the vein in the first place.

Per other discussions, they're now using their own technology only for HSV-1 tests, which are PCR based and therefore require only a tiny sample, and the above sort of contamination would not be an issue.

Agree, and to your point I have read and watched most of the latest interviews given by Holmes. In no place she claims that Theranos is using blood collected in nanotainers (non-venous blood or finger prick blood) for assay (component detection) using traditional diagnostic equipments.

Theranos uses non-venous blood for their proprietary tests (currently only 1 test) and rest 99.9% (figuratively speaking) tests uses venous blood in traditional diagnostic equipment.

Back to the same question - If almost all of their tests are done using traditional equipments, then why $9B valuation?

Seem overvalued. But valuations are on future value. So if they could get their test through the mill, what would they be worth? Multiply by your estimate of their chances. Probably << $9B

I think you hit the nail on the head. The point is that at that valuation, with that much investor money coming in, there's no excuse for them not to have met the minimum barrier to entry in the area they chose, which is essentially compliance with federal regulations.

Are the participants in Theranos' 2014 private equity sale even publicly known? The list could be a source of amusement for many. Most amusing outcome would be if DFJ, Ellison, et al exited in that round.

Document numbers? Release dates? This seems more like red tape and nitpicking rather than a substantive challenge to a technology. It's stories like these that guarantee I would never try to start a company in biotech or recommend a friend join one.

Unit testing? Source control? Continuous integration? Code documentation? Seems like red tape and nitpicking rather than substantive coding. It's stories like that that guarantee that I would never try to start a company in software or recommend a friend join one.

Just because standard practice seems unnecessary to you doesn't mean that it doesn't exist for good reasons. To anyone in biotech or familiar with what can go wrong with physical products, your complaints come across as ignorant.

I've worked at successful companies which did a terrible job at one or another of those, and the customers didn't complain (or die). You can get away with a lot more in software than you can in other disciplines, because in software you can incorporate negative feedback relatively quickly.

The high flying startups you know didn't go to market with the engineering processes you mention.

The FDA quality system regulations seem like nitpicking on their face, and they can be implemented in such a way that is very pedantic, but at their core this is really about an engineering process that is controlled and validated, which isn't, you know, unreasonable if you're talking about products that directly impact people's health.

Document numbers and release dates go toward ensuring everyone at the company is working with the same information. Most computer science students learn about the failure of the Mars Climate Orbiter [1] because one team was using feet and the other meters. Guarding against issues like those is exactly what FDA is trying to do.

1: http://articles.latimes.com/1999/oct/01/news/mn-17288

The FDA is a federal bureaucracy, and the "picky" requirements are the usual and customary mode of operation. Documentation of details is important in order to conform to federal law which the FDA implements in its domains.

Having been involved in several pharmaceutical studies, I can assure you the level of documentation detail demanded is extremely picky. Medical laboratory rules are similarly strict. They have to be given the risks to patients if not closely regulated.

The essence of the requirements are thoroughness and consistency. Medical laboratory measurement devices have to be carefully calibrated in order to be predictable, a lot rides on the accuracy of results. Without conforming documentation of calibration and testing procedures, it's impossible to trust lab test results, which is untenable.

It's not "just red tape", if true that compliance with standards is arduous. I have no trouble understanding reasons people might avoid getting involved in biotech enterprises, or recommending it for others.

TBH the FDA is the least of the various hurdles to get a great biotech going. The FDA has (surprise) largely been spot on. Theranos is just a fraud. If you are not risk tolerant that's fine, it's a reasonable position. But, there are plenty of people who are willing to actually get outside the anonymous video chat / messaging whatever echo chamber and make real inventions happen in the world that make a real difference to real people.

>It's stories like these that guarantee I would never try to start a company in biotech or recommend a friend join one.

Good. We're in mutual agreement. Because I or my friends don't want people working on matters of life and death who don't care about "red tape" or "document numbers".

That's all a part of design control, and it is common among regulated industries (US and abroad). It's not hard to do, it just takes some diligence.

As far as the biotech industry, the funding situation is awful, especially for early companies, and that is a fine reason to avoid this area. Needing to "say what you do and do what you say" doesn't strike me as that much of an obstacle in comparison.

when you start a company your approach should be beyond reproach. red tape and nitpicking are things to welcome as part of a high quality bar - as annoying as they maybe - not to write off as inconvenient annoyances.

when you have your shit together in the most basic ways, you simply won't have this level of problem.

Name a successful startup that was beyond reproach.

most people try this tbh. its common sense. its why you do things like register your company, pay your taxes, get legal advice. its exceptionally normal behaviour until recently it seems...

we used to call people who didn't do this 'rogue traders' and the like. they were looked down on like scum instead of looked up to as role models.

I'm so cynical that I don't trust the FDA to not be looking out for established medical industry players.

I'm not saying conspiratorial that this is all BS but rather that I'm surprised there isn't some more skepticism considering disruption of industries this big and regulated will never go easily.

I understand that folks believe the cult of disruption has blinded everyone in this case and think that may well be true. still these issues highlighted by the FDA sounds to me like any scrappy startup which has to cut some corners. some of you will claim the sky is falling but I don't buy it. if there's nothing to their tech then they will indeed collapse but these papers are not enough to make that clear

is that the same FDA involved in this: http://science.slashdot.org/story/15/10/27/037234/the-popula... ?

And this: http://www.cbsnews.com/news/fda-corruption-letter-authentica...

And this: http://seattleorganicrestaurants.com/vegan-whole-food/instit...

And this: http://ceramics.org/ceramic-tech-today/biomaterials/gsk-drop...

And this: https://reason.com/blog/2013/11/25/fda-shuts-down-23andme-ou...

Why anyone's first inclination would be to trust the FDA is beyond me. In fact, this whole story has a fishy Denmark smell to it. Why for instance is all dissenting opinion being down voted?

Because some people understand why the FDA shut down 23andme and approve of this decision.

Maybe if you would have linked the official statements of both organizations rather than a "reason.com" blog link with opinionated bullshitting, then your point might stand some scrutiny.

This and the fact that things have now shifted [1] because even if that doesn't fill your sensational hunger, no, the FDA is not on a crazy trip to stop personal genomics.

[1] http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/uc...

"Because some people understand why the FDA shut down 23andme"

Oh, do you? How about you explain it then.

Reply on news.ycombinator.com